There goes the neighborhood: Performance degradation due to nearby jobs
There goes the neighborhood: Performance degradation due to nearby jobs
复制标题
DOI：
10.1145/2503210.2503247
复制
发表时间：
2013-11
期刊：
影响因子：
--
通讯作者：
A. Bhatele;K. Mohror;S. Langer;Katherine E. Isaacs
中科院分区：
文献类型：
--
作者：
A. Bhatele;K. Mohror;S. Langer;Katherine E. Isaacs
可预测的性能对于理解和减轻应用程序性能问题非常重要；量化源代码、编译器或系统软件变更的影响；估计批处理作业所需的时间；确定提案的分配请求。我们的实验表明，在Cray XE系统上，通信密集型并行应用程序的执行时间比观察到的平均性能快28%到慢41%。另一方面，蓝色基因系统没有表现出明显的运行差异。在本文中，我们将重点放在Cray机器上，并研究性能变化的潜在原因，如操作系统抖动、分配分区的形状以及来自共享相同网络链接的其他作业的干扰。减少这种可变性可以提高计算机中心的总体吞吐量并节省能源成本。
Predictable performance is important for understanding and alleviating application performance issues; quantifying the effects of source code, compiler, or system software changes; estimating the time required for batch jobs; and determining the allocation requests for proposals. Our experiments show that on a Cray XE system, the execution time of a communication-heavy parallel application ranges from 28% faster to 41% slower than the average observed performance. Blue Gene systems, on the other hand, demonstrate no noticeable run-to-run variability. In this paper, we focus on Cray machines and investigate potential causes for performance variability such as OS jitter, shape of the allocated partition, and interference from other jobs sharing the same network links. Reducing such variability could improve overall throughput at a computer center and save energy costs.