一句话：对共享多租户平台上深度学习作业因代码/脚本缺陷而失败的实证研究。
论文关注企业在共享、多租户平台上提交和运行深度学习程序时，部分作业在长时间执行后因代码或脚本缺陷而失败的问题。这类失败会降低开发生产力，并浪费 GPU、存储和网络 I/O 等昂贵资源。作者针对深度学习作业的程序失败开展实证研究，旨在从实际平台运行视角刻画该问题的表现与影响。摘要未展开具体数据来源、失败分类或统计结果，但明确将深度学习作业失败与代码/脚本缺陷、长时执行及资源共享环境联系起来。该工作有助于后续改进深度学习作业的可靠性、开发效率与资源利用率。
Deep learning has made significant achievements in many application areas. To train and test models more efficiently, enterprise developers submit and run their deep learning programs on a shared, multi-tenant platform. However, some of the programs fail after a long execution time due to code/script defects, which reduces the development productivity and wastes expensive resources such as GPU, storage, and network I/O.
DOI 原文 ·
@inproceedings{paperbot5679,
title = {An empirical study on program failures of deep learning jobs},
author = {Ru Zhang and Wencong Xiao and Hongyu Zhang and Yu Liu and Haoxiang Lin and Mao Yang},
booktitle = {ICSE},
year = {2020},
doi = {10.1145/3377811.3380362}
}