计算机工程

• 体系结构与软件技术 • 上一篇    下一篇

求解大规模三对角线性方程组的GaBP并行算法

陈振武,郑汉垣,兰添才,曾志宏   

  1. (龙岩学院 信息工程学院,福建 龙岩 364012)
  • 收稿日期:2015-10-12 出版日期:2016-10-15 发布日期:2016-10-15
  • 作者简介:陈振武(1963—),男,实验师,主研方向为计算机体系结构;郑汉垣,副教授、博士;兰添才,副教授;曾志宏,讲师、硕士。
  • 基金项目:
    福建省自然科学基金资助项目(2015J01587);福建省教育厅A类基金资助项目(JA09229);福建省科技厅高校基金资助项目(JK2010056);龙岩学院服务海西基金资助项目(JB10160,LYXY2011067)。

Parallel GaBP Algorithm for Solving Large-scale Tridiagonal Linear Equations

CHEN Zhenwu,ZHENG Hanyuan,LAN Tiancai,ZENG Zhihong   

  1. (School of Information Engineering,Longyan University,Longyan,Fujian 364012,China)
  • Received:2015-10-12 Online:2016-10-15 Published:2016-10-15

摘要: 根据大规模三对角线性方程组求解的特性,结合消息传递接口和开放多处理模型,设计分布式共享内存环境下求解大规模三对角线性方程组的GaBP并行算法。在Intel Xeon E5-2650并行计算集群环境上进行数值实验,结果表明,与基于消息传递接口的GaBP并行算法相比,该算法具有更高的加速比和更好的可扩展性,能充分发挥集群系统的综合计算性能,提高大规模三对角线性方程组的求解速度。

关键词: 三对角线性方程组, GaBP算法, 并行算法, 加速比, 可扩展性

Abstract: According to the solution characteristics of large-scale tridiagonal linear equations,based on distributed shared memory environment of mixing Message Passing Interface(MPI) with Open Multi-processing(OpenMP),this paper designs a parallel GaBP algorithm to solve large-scale tridiagonal linear equations.Tests are done through parallel computing cluster environment of Intel Xeon E5-2650,and the results show that the proposed algorithm has higher speed up ratio and better scalability,compared with parallel GaBP algorithm based on MPI.It can take fuller advantage of the overall computing performance of cluster system and improve the speed of solving the large-scale tridiagonal linear equations.

Key words: tridiagonal linear equations, Gaussian Belief Propagation(GaBP) algorithm, parallel algorithm, speed up ratio, scalability

中图分类号: