update README
Browse files
README.md
CHANGED
@@ -5,4 +5,86 @@ task_categories:
|
|
5 |
pretty_name: cloud
|
6 |
size_categories:
|
7 |
- 100M<n<1B
|
8 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
pretty_name: cloud
|
6 |
size_categories:
|
7 |
- 100M<n<1B
|
8 |
+
---
|
9 |
+
|
10 |
+
# Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain
|
11 |
+
[Paper](https://arxiv.org/abs/2310.05063) | [Code](https://github.com/SalesforceAIResearch/pretrain-time-series-cloudops)
|
12 |
+
|
13 |
+
Datasets accompanying the paper "Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain".
|
14 |
+
|
15 |
+
```python
|
16 |
+
from datasets import load_dataset
|
17 |
+
|
18 |
+
dataset = load_dataset('Salesforce/cloudops_tsf', 'azure_vm_traces_2017')
|
19 |
+
```
|
20 |
+
|
21 |
+
### azure_vm_traces_2017
|
22 |
+
```python
|
23 |
+
DatasetDict({
|
24 |
+
train_test: Dataset({
|
25 |
+
features: ['start', 'target', 'item_id', 'feat_static_cat', 'feat_static_real', 'past_feat_dynamic_real'],
|
26 |
+
num_rows: 17568
|
27 |
+
})
|
28 |
+
pretrain: Dataset({
|
29 |
+
features: ['start', 'target', 'item_id', 'feat_static_cat', 'feat_static_real', 'past_feat_dynamic_real'],
|
30 |
+
num_rows: 159472
|
31 |
+
})
|
32 |
+
})
|
33 |
+
```
|
34 |
+
|
35 |
+
### borg_cluster_data_2011
|
36 |
+
```python
|
37 |
+
DatasetDict({
|
38 |
+
train_test: Dataset({
|
39 |
+
features: ['start', 'target', 'item_id', 'feat_static_cat', 'past_feat_dynamic_real'],
|
40 |
+
num_rows: 11117
|
41 |
+
})
|
42 |
+
pretrain: Dataset({
|
43 |
+
features: ['start', 'target', 'item_id', 'feat_static_cat', 'past_feat_dynamic_real'],
|
44 |
+
num_rows: 143386
|
45 |
+
})
|
46 |
+
})
|
47 |
+
```
|
48 |
+
|
49 |
+
### alibaba_cluster_trace_2018
|
50 |
+
```python
|
51 |
+
DatasetDict({
|
52 |
+
train_test: Dataset({
|
53 |
+
features: ['start', 'target', 'item_id', 'feat_static_cat', 'past_feat_dynamic_real'],
|
54 |
+
num_rows: 6048
|
55 |
+
})
|
56 |
+
pretrain: Dataset({
|
57 |
+
features: ['start', 'target', 'item_id', 'feat_static_cat', 'past_feat_dynamic_real'],
|
58 |
+
num_rows: 58409
|
59 |
+
})
|
60 |
+
})
|
61 |
+
```
|
62 |
+
|
63 |
+
## Acknowledgements
|
64 |
+
The datasets were processed from the following original sources. Please cite the original sources if you use the datasets.
|
65 |
+
* Azure VM Traces 2017
|
66 |
+
* Bianchini. Resource central: Understanding and predicting workloads for improved resource
|
67 |
+
management in large cloud platforms. In Proceedings of the 26th Symposium on Operating Systems
|
68 |
+
Principles, pp. 153–167, 2017.
|
69 |
+
* https://github.com/Azure/AzurePublicDataset
|
70 |
+
|
71 |
+
* Borg Cluster Data 2011
|
72 |
+
* John Wilkes. More Google cluster data. Google research blog, November 2011. Posted at http:
|
73 |
+
//googleresearch.blogspot.com/2011/11/more-google-cluster-data.html.
|
74 |
+
* https://github.com/google/cluster-data
|
75 |
+
|
76 |
+
* Alibaba Cluster Trace 2018
|
77 |
+
* Jing Guo, Zihao Chang, Sa Wang, Haiyang Ding, Yihui Feng, Liang Mao, and Yungang Bao. Who
|
78 |
+
limits the resource efficiency of my datacenter: An analysis of alibaba datacenter traces. In
|
79 |
+
Proceedings of the International Symposium on Quality of Service, pp. 1–10, 2019.
|
80 |
+
* https://github.com/alibaba/clusterdata
|
81 |
+
|
82 |
+
## Citation
|
83 |
+
```
|
84 |
+
@article{woo2023pushing,
|
85 |
+
title={Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain},
|
86 |
+
author={Woo, Gerald and Liu, Chenghao and Kumar, Akshat and Sahoo, Doyen},
|
87 |
+
journal={arXiv preprint arXiv:2310.05063},
|
88 |
+
year={2023}
|
89 |
+
}
|
90 |
+
```
|