arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.03632cs.DC
收藏

提高多核NUMA系统中子截面查找程序的可扩展性

Improving the scalabiliy of neutron cross-section lookup codes on multicore NUMA system

Kazutomo Yoshii, John Tramm, Andrew Siegel, Pete Beckman

摘要:

我们使用的XSBench代理应用程序,内存密集型的OpenMP程序,探索一个流行的蒙特卡罗(MC)反应堆物理基准的非均匀内存访问(NUMA)系统的节点上的可扩展性退化的来源。作为背景,我们介绍了XSBench的详细信息,这是一个用于完整MC模拟的性能抽象“代理应用程序”,以及Linux内核的内部设计。我们解释了内核内部的物理内存分配如何影响XSBench的多核可伸缩性。在一个16核双插槽NUMA测试平台上,扩展效率从未优化的70%提高到优化的95%,优化版本的能耗比未优化版本低25%。除了NUMA优化之外,我们还评估了XSBench的页面大小优化,并观察到与未优化的页面大小优化相比,性能提高了1.5倍。

英文摘要:

We use the XSBench proxy application, a memory-intensive OpenMP program, to explore the source of on-node scalability degradation of a popular Monte Carlo (MC) reactor physics benchmark on non-uniform memory access (NUMA) systems. As background, we present the details of XSBench, a performance abstraction "proxy app" for the full MC simulation, as well as the internal design of the Linux kernel. We explain how the physical memory allocation inside the kernel affects the multicore scalability of XSBench. On a sixteen-core, two-socket NUMA testbed, the scaling efficiency is improved from a nonoptimized 70% to an optimized 95%, and the optimized version consumes 25% less energy than does the nonoptimized version. In addition to the NUMA optimization we evaluate a page-size optimization to XSBench and observe a 1.5x performance improvement, compared with a nonoptimized one.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑