High Performance Computing - ISC High Performance 2019 International Workshops, Frankfurt, Germany, June 16-20, 2019, Revised Selected Papers

High Performance Computing - ISC High Performance 2019 International Workshops, Frankfurt, Germany, June 16-20, 2019, Revised Selected Papers
复制标题

高性能计算 - ISC 高性能 2019 国际研讨会,德国法兰克福,2019 年 6 月 16-20 日,修订后的精选论文

DOI:
10.1007/978-3-030-34356-9_33
复制
发表时间:
2019
期刊:
--
影响因子:
--
通讯作者:
Bombace N
Bombace N
中科院分区:
--
文献类型:
--
作者:
Bombace N

文献摘要

相似文献

由于非确定性并行归约和非关联浮点运算的综合影响,并行计算本质上是不可再现的。文献中提出了不同的策略来缓解或完全消除这个问题,但是目前还没有关于关联浮点运算对大规模应用的性能影响的研究。在这项工作中,我们使用 MiniFE 中的分箱双精度数实现关联运算,并在 Cirrus 和 Fulhame 这两个最先进的 HPC 系统上执行各种性能测试。
Parallel computations are intrinsically non-reproducible, due to a combined effect of non-deterministic parallel reductions and non-associative floating point operations. Different strategies have been proposed in literature to alleviate this issue or eliminate it altogether, however at present there is no study on the performance impact of associative floating point operations on large scale applications. In this work, we implement associative operations using binned doubles in MiniFE, and perform various performance tests on Cirrus and Fulhame, two state-of-the-art HPC systems.