基于MPI的高精度归约函数设计与实现  

Design and implementation of a high-precision reduction function based on MPI

作　　者：何康 黄春[1] 姜浩[1] 谷同祥[2] 齐进[2] 刘杰[1,3,4] HE Kang;HUANG Chun;JIANG Hao;GU Tong-xiang;QI Jin;LIU Jie(College of Computer Science and Technology,National University of Defense Technology,Changsha 410073;Institute of Applied Physics and Computational Mathematics,Beijing 100000;Science and Technology on Parallel and Distributed Processing Laboratory,National University of Defense Technology,Changsha 410073;Laboratory of Software Engineering for Complex Systems,National University of Defense Technology,Changsha 410073,China)

机构地区：[1]国防科技大学计算机学院,湖南长沙410073 [2]北京应用物理与计算数学研究所,北京100000 [3]国防科技大学并行与分布处理重点实验室,湖南长沙410073 [4]国防科技大学复杂系统软件工程湖南省重点实验室,湖南长沙410073

出　　处：《计算机工程与科学》2021年第4期594-602,共9页Computer Engineering & Science

基　　金：重点研发计划(2018YFB0204301);国家自然科学基金(61907034);科学挑战专题资助项目(TZ2016002);湖南省自然科学基金(2018JJ3616)。

摘　　要：随着科学工程计算大规模、高维数和长时程的特性越来越显著,浮点舍入误差的累积效应往往使得计算结果不可信,提高计算精度成为了并行计算领域研究的热点之一。基于MPICH3框架,采用无误差变换技术构建新的数据格式和相应运算操作符,设计了高精度归约函数MPI_ACCU_REDUCE,实现了高精度的求和、求积和求L2范数3种MPI归约运算。数值实验结果表明,提出的3种高精度归约运算有效提高了数值计算的精度。As the characteristics of large scale, high dimension, and long time of scientific and engineering computing become more and more obvious, the cumulative effect of floating-point rounding errors often makes calculation results unreliable. Improving the computing accuracy has become one of the hot spots in the field of parallel computing. Based on the MPICH3 framework, this paper uses error-free transformation to construct a new data format and corresponding operators, and designs a high-precision reduction function MPI_ACCU_REDUCE, which realizes three types of high-precision MPI reduction operations such as summation, production and L2 norm. Numerical experiment shows that the proposed MPI_ACCU_REDUCE function with the three types of high-precision reduction operations effectively improve the accuracy of numerical calculations.

关 键 词：MPI 高精度计算 归约操作 无误差变换 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]