Automating instrumentation choices for performance problems in distributed applications with VAIF

Automating instrumentation choices for performance problems in distributed applications with VAIF
复制标题

使用 VAIF 自动选择仪器来解决分布式应用程序中的性能问题

DOI:
10.1145/3472883.3487000
复制
发表时间:
2021
期刊:
Proceedings of the ACM Symposium on Cloud Computing
影响因子:
--
通讯作者:
Sambasivan, Raja R.
Sambasivan, Raja R.
中科院分区:
--
文献类型:
--
作者:
Toslali, Mert;Ates, Emre;Ellis, Alex;Zhang, Zhaoqi;Huye, Darby;Liu, Lan;Puterman, Samantha;Coskun, Ayse K.;Sambasivan, Raja R.

文献摘要

参考文献

被引文献

相似文献

开发人员使用日志来诊断分布式应用程序中的性能问题。然而,很难先验地知道哪里需要日志,以及需要日志中的哪些信息来帮助诊断将来可能发生的问题。我们提出了方差驱动的自动化仪器框架(VAIF),它与分布式应用程序一起运行。为了响应新观察到的性能问题,VAIF自动搜索可能的检测选项空间,以启用帮助诊断它们所需的日志。为了工作,VAIF将分布式跟踪(一种增强的日志记录形式)与如何在请求跟踪的关键路径部分上分解响应时间方差的洞察相结合。我们通过使用VAIF来评估OpenStack和HDFS中的性能问题。我们表明,VAIF可以本地化的问题,缓慢的代码路径,资源争用,有问题的第三方代码,而使只有3-34%的总跟踪仪器。
Developers use logs to diagnose performance problems in distributed applications. However, it is difficult to know a priori where logs are needed and what information in them is needed to help diagnose problems that may occur in the future. We present the Variance-driven Automated Instrumentation Framework (VAIF), which runs alongside distributed applications. In response to newly-observed performance problems, VAIF automatically searches the space of possible instrumentation choices to enable the logs needed to help diagnose them. To work, VAIF combines distributed tracing (an enhanced form of logging) with insights about how response-time variance can be decomposed on the critical-path portions of requests' traces. We evaluate VAIF by using it to localize performance problems in OpenStack and HDFS. We show that VAIF can localize problems related to slow code paths, resource contention, and problematic third-party code while enabling only 3-34% of the total tracing instrumentation.
DOI: --
发表时间: 2010
期刊: --
影响因子: --
作者:
B. Sigelman;L. Barroso;M. Burrows;Patrick Stephenson;Manoj Plakal;Donald Beaver;Saul Jaspan;C. Shanbh
通讯作者: B. Sigelman;L. Barroso;M. Burrows;Patrick Stephenson;Manoj Plakal;Donald Beaver;Saul Jaspan;C. Shanbh
大型系统中的根本原因定位
DOI: --
发表时间: 2005
期刊:
影响因子: --
作者:
Emre Kıcıman;L. Subramanian
通讯作者: L. Subramanian