Datasets:

eastwind

/

finweb_sentences_10BT_40_char

like 0

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_27

Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_28

By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_29

A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_30

I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_31

A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_32

A transaction is initiated by a single request and may take many individual bus operations.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_33

A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_34

Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_35

A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_36

The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_37

The one connected to north bridge, witch is the I/O controller hub.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_38

Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_39

Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_40

- Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_41

Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_42

- Make the transfer of data interrupt friven.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_43

- direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_44

I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_45

Transaction processing applications typically have both response time requirements and throughput performance.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_46

Performance measure of I/Os per unit time, such as reads per second.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_47

Performance measure of bytes per unit time, such as GB/sec.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_48

Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_49

- Configure this component to sustain the required bandwidth.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_50

- Determine the requirements for the rest of the system and configure them to support this bandwidth.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_51

Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_52

No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_53

Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_54

Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_55

Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_56

The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_57

Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_58

Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_59

P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_60

The second check block allows recovery from a second failure.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_61

Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_62

MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_63

The annual failure rate (AFR) is a more useful measure.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_64

Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_65

- Generally cannot use 100% of any computer resource.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_66

- The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_67

OSs are the best place to schedule disk accesses.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_68

Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_69

Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_70

The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations.

Dependability, Reliability, and Availability Dependability is the quality of delivered service such that reliance can justifiably be placed on this service. Reliability is a measure of the continuous service accomplishment(or, equivalently, of the time to failure) from a reference point. mean time to failure(MTTF) is a reliability measure. annual failure rate(AFR) is the percentage of devices that would be expected to fail in a year for a given MTTF. Service interruption is measured as mean time to repair(MTTR). Mean time between failures(MTBF) is simply the sum of MTTF and MTTR. Availability is a measure of service accomplishment with respect to the alternation between the two states of accomplishment and interruption. Availability = MTTF / MTBF 3 ways to improve MTTF: - Fault avoidance: preventing fault occurrence by construction. - Fault tolerance: using redundancy to allow the service to comply with the service specification despite faults occurring, which applies primarily to hardware faults. - Fault forecasting: predicting the presence and creation of faults, which applies to hardware and software faults, allowing the component to be replaced before it fails. Storage device where data retins its value even when power is removed. One of thousands of concentric circles that makes up the surface of a magnetic disk. One of the segments that make up a track on a magnetic disk; a sector is the smallest amount of information that is read or written on a disk. Originally, all tracks had the same number of sectors and the same number of bits. With the introduction of zone bit recording(ZBR) , disk drives changed to a varying number of sectors per track. Thus increasing the drive capacity. Cylinder referes to all the tracks under the heads at a given point on all surfaces. The time for the process of positioning a read/write head over the proper track on a disk. Also called rotational delay , the time required for the desired sector of a disk to rotate under the read/write head; usually assumed to be half the rotation time. Transfer time is the time to transfer a block of bits. Disk controller usually handles the detailed control of the disk and the transfer between the disk and the memory. Controller time is the overhead the controller imposes in performing an I/O access. NOR flash : storage cell is similiar to a standard NOR gate. Typically used for BIOS memory. NAND flash : offers greater storage density, but memory could only be read and written in blocks as wiring needed for random accesses was removed. Typically used for USB key. To cope with bit wearing out, most NAND flash products include a controller to spread the writes by remaapping blocks that have been written many times to less trodden blocks. Connecting Processors, Memory, and I/O Devices A bus that connects processor and memory and that is short, generally high speed, and matched to the memory system so as to maximize memory-processor bandwidth. By contrast, I/O bus can be lengthy, can have many types of devices connected to them, and often have a wide range in data bandwidth of the devices connected to them. A bus that is designed to allow processors, memory, and I/O devices to coexist on a single bus. I/O buses donot typically interface directly to the memoty but use either a processor-memory or a backplane bus to connect to memory. A sequence of operations over the interconnect that includes a request and may include a response, either of which may caryy data. A transaction is initiated by a single request and may take many individual bus operations. A bus that includes a clock in the control lines and a fixed protocol for communicating that is relative to the clock. Uses a handshaking protocol for coordinating usage rather than a clock; can accommodate a wide variety of devices of differing speeds. A series of steps used to coordinate asynchronous bus transfers in which the sender and receiver proceed to the next step only when both parties agree that the current step has been completed. The I/O Interconnects of the x86 Processors The chip for memory controller hub next to the processor. The one connected to north bridge, witch is the I/O controller hub. Interfacing I/O Devices to the Processor, Memory, and Operating System Giving commands to I/O devices An I/O scheme in which portions of address space are assigned to I/O devices, and reads and writes to those addresses are interpreted as commands to the I/O device. Communicating with the Processor - polling : The process of periodically checking the status of an I/O device to determine the need to service the device. - Interrupt-driven I/O : An I/O scheme that employs interrupts to indicate to the processor that an I/O device needs attention. Transferring the Data between a Device and Memory - Use the processor to transfer data between a device and memory based on polling. - Make the transfer of data interrupt friven. - direct memory access (DMA) : Having the device controller transfer data directly to ro from the memory without involving the processor. I/O Performance Measures A type of application that involves handing small short operations(called transactions) that typically require both I/O and computation. Transaction processing applications typically have both response time requirements and throughput performance. Performance measure of I/Os per unit time, such as reads per second. Performance measure of bytes per unit time, such as GB/sec. Designing and I/O System - Find the weakest link in the I/O system, which will constrain the design. - Configure this component to sustain the required bandwidth. - Determine the requirements for the rest of the system and configure them to support this bandwidth. Parallelism and I/O: Redundant Arrays of Inexpensive Disks redundant arrays of inexpensive disks (RAID) An organization of disks that uses an array of small and inexpensive disks so as to increase both performance and reliability. No Redundancy (RAID 0) Simply spreading data over multiple disks, called striping. Mirroring (RAID 1) Writing the identical data to multiple disks to increase data availability. Error Detecting and Correcting Code (RAID 2) RAID 2 borrows an error detection and correction scheme most often used for memories. Bit-Interleaved Parit (RAID 3) protection group is the group of data disks or blocks that share a common check disk or block. The cost of higher availability can be reduced to 1/n, where n is the number of disks in a protection group. Block-Interleaved Parity (RAID 4) The parity is stored as blocks and associated with a set of data blocks. Distributed Block-Interleaved Parity (RAID 5) To fix the parity-write bottleneck, the parity information can be spread throughout all the disks so that there is no single bottleneck for writes. P+R Redundancy (RAID 6) When a single failure correction is not sufficient, parity can be generalized to have a second calculation over the data and another check disk of information. The second check block allows recovery from a second failure. Fallacies and Pitfalls Fallacy: The rated mean time to failure of disks is almost 140 years, so disks practically never fail. MTTF is calculated by putting thousands of disks in a room, run them for a few months, and count the number that fail. The annual failure rate (AFR) is a more useful measure. Fallacy: A GB/sec interconnect can transfer 1 GB of data in 1 second. - Generally cannot use 100% of any computer resource. - The definition of a GB of storage ($2^30$) and a GB persecond of bandwidth ($10^9$) donot agree. OSs are the best place to schedule disk accesses. Since the disk knows the actual mapping of the logical addresses onto a physical geometry of sectors, tracks, and surfaes, it can reduce the rotational and seek latencies by rescheduling. Pitfall: Using the peak transfer rate of a portion of the I/O system to make performance projections or performance comparisons. The peak performance is based on unrealistic assumptions about the system or are unattainble because of other system limitations. 转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>

CC-MAIN-2018-13

http://harttle.land/2014/02/23/computer-design-io.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:4f2203bb-f3c2-4d1b-ab0d-978125506e16>_71

转载请注明来源： http://harttle.land/2014/02/23/computer-design-io.html 欢迎对文章中的引用来源进行考证，欢迎指出任何有错误或不够清晰的表达。可以在下面评论区评论（可能需要在能访问 disqus 服务的网络），也可以邮件至 email@example.com。

By Angela P. Dodson In November, we honor the indigenous people of the United States. It is National Native American Heritage Month. It has been proclaimed by law since 1990 when President George H. W. Bush declared the month of November as National American Indian Heritage Month. It later became known by its current name. This month was chosen because it represents the end of harvest time and includes the national celebration of Thanksgiving. According to the U.S. Census Bureau, in 2010 more than 2.9 million people identify as American Indian and Alaska Natives, and about 2.3 million more report native heritage in combination with one or more other races, totaling about 1.6 percent of the nation’s population. About 500 tribes live in the U.S. The observance this month offers opportunities to attend programs, participate in activities and read about the customs and contributions of the first people of this land. Head to our American Indian/Alaskan Native page for resources and information about getting involved on PC's campus.

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>

CC-MAIN-2018-13

https://www.portervillecollege.edu/node/5476

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>_0

By Angela P. Dodson In November, we honor the indigenous people of the United States.

By Angela P. Dodson In November, we honor the indigenous people of the United States. It is National Native American Heritage Month. It has been proclaimed by law since 1990 when President George H. W. Bush declared the month of November as National American Indian Heritage Month. It later became known by its current name. This month was chosen because it represents the end of harvest time and includes the national celebration of Thanksgiving. According to the U.S. Census Bureau, in 2010 more than 2.9 million people identify as American Indian and Alaska Natives, and about 2.3 million more report native heritage in combination with one or more other races, totaling about 1.6 percent of the nation’s population. About 500 tribes live in the U.S. The observance this month offers opportunities to attend programs, participate in activities and read about the customs and contributions of the first people of this land. Head to our American Indian/Alaskan Native page for resources and information about getting involved on PC's campus.

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>

CC-MAIN-2018-13

https://www.portervillecollege.edu/node/5476

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>_1

It is National Native American Heritage Month.

By Angela P. Dodson In November, we honor the indigenous people of the United States. It is National Native American Heritage Month. It has been proclaimed by law since 1990 when President George H. W. Bush declared the month of November as National American Indian Heritage Month. It later became known by its current name. This month was chosen because it represents the end of harvest time and includes the national celebration of Thanksgiving. According to the U.S. Census Bureau, in 2010 more than 2.9 million people identify as American Indian and Alaska Natives, and about 2.3 million more report native heritage in combination with one or more other races, totaling about 1.6 percent of the nation’s population. About 500 tribes live in the U.S. The observance this month offers opportunities to attend programs, participate in activities and read about the customs and contributions of the first people of this land. Head to our American Indian/Alaskan Native page for resources and information about getting involved on PC's campus.

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>

CC-MAIN-2018-13

https://www.portervillecollege.edu/node/5476

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>_2

It has been proclaimed by law since 1990 when President George H. W. Bush declared the month of November as National American Indian Heritage Month.

By Angela P. Dodson In November, we honor the indigenous people of the United States. It is National Native American Heritage Month. It has been proclaimed by law since 1990 when President George H. W. Bush declared the month of November as National American Indian Heritage Month. It later became known by its current name. This month was chosen because it represents the end of harvest time and includes the national celebration of Thanksgiving. According to the U.S. Census Bureau, in 2010 more than 2.9 million people identify as American Indian and Alaska Natives, and about 2.3 million more report native heritage in combination with one or more other races, totaling about 1.6 percent of the nation’s population. About 500 tribes live in the U.S. The observance this month offers opportunities to attend programs, participate in activities and read about the customs and contributions of the first people of this land. Head to our American Indian/Alaskan Native page for resources and information about getting involved on PC's campus.

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>

CC-MAIN-2018-13

https://www.portervillecollege.edu/node/5476

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>_3

It later became known by its current name.

By Angela P. Dodson In November, we honor the indigenous people of the United States. It is National Native American Heritage Month. It has been proclaimed by law since 1990 when President George H. W. Bush declared the month of November as National American Indian Heritage Month. It later became known by its current name. This month was chosen because it represents the end of harvest time and includes the national celebration of Thanksgiving. According to the U.S. Census Bureau, in 2010 more than 2.9 million people identify as American Indian and Alaska Natives, and about 2.3 million more report native heritage in combination with one or more other races, totaling about 1.6 percent of the nation’s population. About 500 tribes live in the U.S. The observance this month offers opportunities to attend programs, participate in activities and read about the customs and contributions of the first people of this land. Head to our American Indian/Alaskan Native page for resources and information about getting involved on PC's campus.

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>

CC-MAIN-2018-13

https://www.portervillecollege.edu/node/5476

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>_4

This month was chosen because it represents the end of harvest time and includes the national celebration of Thanksgiving.

By Angela P. Dodson In November, we honor the indigenous people of the United States. It is National Native American Heritage Month. It has been proclaimed by law since 1990 when President George H. W. Bush declared the month of November as National American Indian Heritage Month. It later became known by its current name. This month was chosen because it represents the end of harvest time and includes the national celebration of Thanksgiving. According to the U.S. Census Bureau, in 2010 more than 2.9 million people identify as American Indian and Alaska Natives, and about 2.3 million more report native heritage in combination with one or more other races, totaling about 1.6 percent of the nation’s population. About 500 tribes live in the U.S. The observance this month offers opportunities to attend programs, participate in activities and read about the customs and contributions of the first people of this land. Head to our American Indian/Alaskan Native page for resources and information about getting involved on PC's campus.

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>

CC-MAIN-2018-13

https://www.portervillecollege.edu/node/5476

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>_5

According to the U.S. Census Bureau, in 2010 more than 2.9 million people identify as American Indian and Alaska Natives, and about 2.3 million more report native heritage in combination with one or more other races, totaling about 1.6 percent of the nation’s population.

By Angela P. Dodson In November, we honor the indigenous people of the United States. It is National Native American Heritage Month. It has been proclaimed by law since 1990 when President George H. W. Bush declared the month of November as National American Indian Heritage Month. It later became known by its current name. This month was chosen because it represents the end of harvest time and includes the national celebration of Thanksgiving. According to the U.S. Census Bureau, in 2010 more than 2.9 million people identify as American Indian and Alaska Natives, and about 2.3 million more report native heritage in combination with one or more other races, totaling about 1.6 percent of the nation’s population. About 500 tribes live in the U.S. The observance this month offers opportunities to attend programs, participate in activities and read about the customs and contributions of the first people of this land. Head to our American Indian/Alaskan Native page for resources and information about getting involved on PC's campus.

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>

CC-MAIN-2018-13

https://www.portervillecollege.edu/node/5476

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>_7

The observance this month offers opportunities to attend programs, participate in activities and read about the customs and contributions of the first people of this land.

By Angela P. Dodson In November, we honor the indigenous people of the United States. It is National Native American Heritage Month. It has been proclaimed by law since 1990 when President George H. W. Bush declared the month of November as National American Indian Heritage Month. It later became known by its current name. This month was chosen because it represents the end of harvest time and includes the national celebration of Thanksgiving. According to the U.S. Census Bureau, in 2010 more than 2.9 million people identify as American Indian and Alaska Natives, and about 2.3 million more report native heritage in combination with one or more other races, totaling about 1.6 percent of the nation’s population. About 500 tribes live in the U.S. The observance this month offers opportunities to attend programs, participate in activities and read about the customs and contributions of the first people of this land. Head to our American Indian/Alaskan Native page for resources and information about getting involved on PC's campus.

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>

CC-MAIN-2018-13

https://www.portervillecollege.edu/node/5476

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257649683.30/warc/CC-MAIN-20180324034649-20180324054649-00510.warc.gz

en

<urn:uuid:61a2ab41-76af-4997-9395-510579868ed1>_8

Head to our American Indian/Alaskan Native page for resources and information about getting involved on PC's campus.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_0

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_1

It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_2

Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_3

The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_4

Therefore, a building receives one of the four levels of certification i.e.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_6

Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_7

One great way of generating points is by the use of recycled materials.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_8

Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_9

It not only increases the points a building can receive but also condenses energy bills.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_10

Home renovation is an excellent way of improving the resale value of a home.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_11

There is a couple of ideas homeowners and designers can use to think green when renovating their homes.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_12

In the case of repainting a home, use of environmental-friendly latex paint is ideal.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_13

All the leftover paints should be left to dry before disposing of.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_14

Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_15

Going green is not only considering the design of a building.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_16

It also means minding the pocket as well.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_17

Installation of solar panels is becoming a popular trend in most green homes.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_18

The sun is in plenty three-quarters of the month.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_19

Solar power can be harnessed and used to heat and light a home in several occasions.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_20

Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_21

For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_22

Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_23

They provide perfect finishes with no toxic fumes or volatile organic compounds.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_24

Cabinets made of straw or wheat board are the most suitable.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_25

Eco-friendly flooring is another splendid idea to consider.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_26

Cork and linoleum make the most sustainable choices for green flooring.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_28

They come in a multitude range of colors and patterns and are resistant to staining.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_29

Such flooring should be obtained from a forest source that uses renewable lumber trends.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_30

When staging a green real estate property for sale in the market, owners must stress the fact that they are green.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_31

It makes a home economically marketable.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_32

Green homes are becoming popular among the eco-friendly folks.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_33

The trend is considerably rising too among the green-minded investors.

As energy costs rise steadily, going green is becoming a growing trend in the real estate industry. It shows that the bid to preserve and salvage the planet is not only limited to automobiles alone. Homeowners have realized the significance of using environmental-friendly products when building and renovating their homes. The U.S. Green Building Council came up with a designation called Leadership in Energy and Environmental Design (LEED) which aims at cutting down energy use and building’s environmental impact. Therefore, a building receives one of the four levels of certification i.e. Certified, Silver, Gold and Platinum. Architects, designers, and organizations can utilize several techniques, trends, and ideas to make a building green while not incurring substantial costs. One great way of generating points is by the use of recycled materials. Altering the direction, a building faces so that natural light can be harnessed via windows or other open spaces is a brilliant idea that does not need any additional cost. It not only increases the points a building can receive but also condenses energy bills. Home renovation is an excellent way of improving the resale value of a home. There is a couple of ideas homeowners and designers can use to think green when renovating their homes. In the case of repainting a home, use of environmental-friendly latex paint is ideal. All the leftover paints should be left to dry before disposing of. Additionally, it is advisable not to dump paint down a drain as it poses and environmental hazard. Going green is not only considering the design of a building. It also means minding the pocket as well. Installation of solar panels is becoming a popular trend in most green homes. The sun is in plenty three-quarters of the month. Solar power can be harnessed and used to heat and light a home in several occasions. Solar panels efficiently convert solar energy into electrical energy that can be used to heat and power electrical appliances in a home. For kitchen and bathroom renovations, going green means installing appliances and fixtures that reduce water wastage. Homeowners are also advised to emphasize on the installation of sustainable kitchen cabinets made of wood certified by the Forest Stewardship Council. They provide perfect finishes with no toxic fumes or volatile organic compounds. Cabinets made of straw or wheat board are the most suitable. Eco-friendly flooring is another splendid idea to consider. Cork and linoleum make the most sustainable choices for green flooring. They are stylish and durable. They come in a multitude range of colors and patterns and are resistant to staining. Such flooring should be obtained from a forest source that uses renewable lumber trends. When staging a green real estate property for sale in the market, owners must stress the fact that they are green. It makes a home economically marketable. Green homes are becoming popular among the eco-friendly folks. The trend is considerably rising too among the green-minded investors. This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>

CC-MAIN-2018-13

http://www.bryanadlam.com/green-home-renovation-ideas/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257651465.90/warc/CC-MAIN-20180324225928-20180325005928-00510.warc.gz

en

<urn:uuid:43528958-a47f-4886-b7cc-a37aaf9411b1>_34

This growing real estate trend and the universal acceptance of green homes implies that it is a profitable investment opportunity.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_1

A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_2

Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_3

The study included data on over 38,000 women who gave birth between 1997 and 2009.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_4

Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_5

The risk of defects was 2 to 3.5 times higher with Paxil and Prozac.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_6

The study did not identify an elevated risk with Zoloft, Celexa or Lexapro.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_7

The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_8

The risk of a heart defect rose from 10 to 24 per 10,000 births.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_9

Reefhuis stated that “the overall risk is still small”.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_10

While research points to risks, women should not stop taking their medication without first contacting their physician.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_11

The study was led by Jennita Reefhuis, an epidemiologist.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_12

“If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR.

Prozac, Paxil Risk Of Birth Defects. A new study published in the British Medical Journal (BMJ) adds to a body of research suggesting that the use of some antidepressants may be associated with a higher risk of birth defects. Researchers at the Centers for Disease Control and Prevention identified an increased risk of defects, including heart abnormalities, abdominal wall abnormalities and missing brain and skull defects, with Prozac and Paxil. The study included data on over 38,000 women who gave birth between 1997 and 2009. Researchers analyzed the rate of birth defects among these children and asked whether mothers took antidepressants in the month before conception or in the first trimester. The risk of defects was 2 to 3.5 times higher with Paxil and Prozac. The study did not identify an elevated risk with Zoloft, Celexa or Lexapro. The absolute risk of anencephaly, or malformation of the brain and skull, increased from 2 to 7 per 10,000 births in children who were exposed to Paxil early in pregnancy. The risk of a heart defect rose from 10 to 24 per 10,000 births. Reefhuis stated that “the overall risk is still small”. While research points to risks, women should not stop taking their medication without first contacting their physician. The study was led by Jennita Reefhuis, an epidemiologist. “If you’re planning to get pregnant, it really is important to talk with your health care provider to see if there are options and if you could choose a safer option among antidepressants,” Reefhuis said, according to NPR. “Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>

CC-MAIN-2018-13

https://www.yourlawyer.com/parker-waichman-llp/new-study-suggests-increased-risk-birth-defects-prozac-paxil/

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645310.34/warc/CC-MAIN-20180317194447-20180317214447-00710.warc.gz

en

<urn:uuid:c7017de9-7e81-4f49-b8c5-00308108ff9f>_13

“Of course, not all antidepressants work the same for everybody, so that’s something that women need to consult with their health care provider about.”

Image Area: The portion of the paper in which ink or toner appears. Image Resolution: The fineness or coarseness of an image as it was digitized, measured as dots-per-inch. Imagesetter: An imaging device specially applied to create type an graphics. Uses raster or vector techniques to expose photographic paper, film, or plate material. Imposition: Laying out of pages in a press form so that they will be in the correct order after the printed sheet is folded. This printed sheet is called a signature. Imprint: To print or image additional copy to a piece that has already been imaged. Independent Graphic: A graphic placed in a publication that is not tied to the text surrounding it. ISDN: Integrated Services Digital Network. A high-bandwidth communications service. ISP: Internet Service Provider. An organization that provides internet access to a consumer.

<urn:uuid:a630f96e-002e-45d8-8c6a-825e37e1dc41>

CC-MAIN-2018-13

http://outwestprinting.com/glossary/i.html

s3://commoncrawl/crawl-data/CC-MAIN-2018-13/segments/1521257645824.5/warc/CC-MAIN-20180318145821-20180318165821-00710.warc.gz

en

<urn:uuid:a630f96e-002e-45d8-8c6a-825e37e1dc41>_0

Image Area: The portion of the paper in which ink or toner appears.