alibaba-edu / dcbrain

dcbrain

dcbrain

We release the following datasets that are collected at Alibaba:

• Hard drive disks (HDDs) (diskdata/): It includes over 200 thousand HDDs in Alibaba Cloud's data centers.

  • Publication: "Large-Scale Disk Failure Prediction(book)."
    Cheng He, Mengling Feng, Patrick P. C. Lee, Pinghui Wang, Shujie Han, Yi Liu.
    PAKDD 2020 Competition and Workshop, AI Ops 2020, February 7 – May 15, 2020, Revised Selected Papers

• Solid-state drives (SSDs) (ssd_open_data/): It includes nearly one million SSDs of 11 drive models from three vendors over a two-year span.

  • Publication: "An In-Depth Study of Correlated Failures in Production SSD-Based Data Centers."
    Shujie Han, Patrick P. C. Lee, Fan Xu, Yi Liu, Cheng He, and Jiongzhou Liu.
    Proceedings of the 19th USENIX Conference on File and Storage Technologies (FAST 2021), February 2021.

• SMART logs of Solid-state drives (SSDs) (ssd_smart_logs/): It includes nearly 500K SSDs of six drive models from three vendors over a two-year span.

  • Publication: "General Feature Selection for Failure Prediction in Large-scale SSD Deployment."
    Fan Xu, Shujie Han, Patrick P. C. Lee, Yi Liu, Cheng He, and Jiongzhou Liu.
    Proceedings of the 51st IEEE/IFIP International Conference on Dependable Systems and Networks (DSN 2021), June 2021.

  • See details about the relationships between two datasets of SSDs.

• DRAM error logs (mcelog) (dramdata/): It includes DRAM errors log, inventory logs, and trouble tickets of server failures due to DRAM errors collected from more than 250K servers over a eight-month span.

  • Publication: "An In-Depth Correlative Study Between DRAM Errors and Server Failures in Production Data Centers."
    Zhinan Cheng, Shujie Han, Patrick P. C. Lee, Xin Li, Jiongzhou Liu, and Zhan Li.
    Proceedings of the 41st International Symposium on Reliable Distributed Systems (SRDS 2022), September 2022.

We release the following prototypes:

• AIHS Prototype (AIHS_prototype/): A prototype for automated intelligent healing in Alibaba Cloud's data centers.

  • Publication: "Automated Intelligent Healing in Cloud-Scale Data Centers."
    Rui Li, Zhinan Cheng, Patrick P. C. Lee, Pinghui Wang, Yi Qiang, Lin Lan, Cheng He, Jinlong Lu, Mian Wang, Xinquan Ding.
    Proceedings of the 40th International Symposium on Reliable Distributed Systems (SRDS 2021), September 2021.

dcbrain's Issues

SMART属性问题

请问，在您的数据表里，r_241代表的是写入数据的总量，那具体的怎么换算成写入量呢？我的理解是他代表自硬盘启用后主机向硬盘写入的数据总量，以4个字节表示，每写入64GB字节作为一个单位。那对于图片的这个数据写入量应该是：
495740215769/4 B=123935053942.25 KB=123935053942.25/1024/1024 GB=118193.6778GB，对吗？

数据标签问题

您好，在您的关于这篇数据集的论文中，有A1-A6、B1-B3、C1-C2这几种类型，但是在数据集中看到在“manufacturer”那一列只显示A，这是什么原因呢？

是否能公开评价指标计算的代码？

请问能否公开 PAKDD2020 阿里巴巴智能运维算法大赛 中评价指标计算的代码？

以便进行更多的算法改进时更好的验证。

https://tianchi.aliyun.com/competition/entrance/231775/information

谢谢！

diskdata 下载链接失效了 ，是否能更新下

关于SMART属性的归一化

您好，请问在smartlog_data_*.csv文件中，smart原始属性是用什么公式进行归一化的呢？谢谢