迭代法解线性方程组的并行算法设计与实现  

Design and Implementation of Parallel Algorithm for Solving Linear Equations by Iterative Method

作　　者：姬进[1] 艾莉[1] 郭警涛[1] 赵君[1] 闫稳[1] Ji Jin;Ai Li;Guo Jingtao;Zhao Jun;Yan Wen(AVIC Computing Technique Research Institute,Xi’an Shaanxi 710065,China)

机构地区：[1]航空工业西安航空计算技术研究所,陕西西安710065

出　　处：《山西电子技术》2022年第4期18-20,23,共4页Shanxi Electronic Technology

摘　　要：本文首先介绍了线性方程组的并行计算和迭代法求解线性方程组,然后重点对雅克比迭代和高斯-赛德尔迭代求解线性方程组的并行计算过程进行了分析,并对两种算法进行了设计和实现,通过测试及性能分析表明雅克比迭代的并行性更好,而高斯-赛德尔迭代的收敛速度更快。This paper first introduces the parallel computing of linear equations and the iterative method to solve linear equations,then focuses on the parallel computing process of Jacobi iteration and Gauss Seidel iteration to solve linear equations,and designs and implements the two algorithms.The test and performance analysis show that Jacobi iteration has better parallelism,and the convergence speed of Gauss Seidel iteration is faster.

关 键 词：线性方程组 迭代法 并行计算 

分 类 号：TP316[自动化与计算机技术—计算机软件与理论]