计算机科学 ›› 2015, Vol. 42 ›› Issue (8): 28-31.

• 目次 • 上一篇    下一篇

关于多核系统并行程序效率的编程因素及其研究

王文义,冉晓龙   

  1. 中原工学院并行处理技术研究所 郑州450007,中原工学院并行处理技术研究所 郑州450007
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受国家自然科学基金项目(61379079),河南省基础与前沿技术研究项目(082300410300)资助

Programming Factors about Efficiency of Parallel Program in Multi-core System and its Research

WANG Wen-yi and RAN Xiao-long   

  • Online:2018-11-14 Published:2018-11-14

摘要: 着重分析了多核架构系统中内存对齐技术与cache利用率等因素对并行程序性能的影响。用共享存储环境OpenMP分析了并行计算量与处理器核心数目之间的关系,通过用MPI编程实现的矩阵相乘的行划分和CANNON算法等实例分析,指出了只有综合考虑了多核系统的结构特征、系统软件、多核编程语言环境以及正确运用算法等,才能设计出高效且能耗又小的并行应用程序。

关键词: 绿色计算,内存对齐,OpenMP,CANNON算法,多核处理器

Abstract: This paper emphatically analyzed the effects of the memory alignment technology,cache using ratio and such as these factors on parallel program performance in the multi-core architecture system.This paper used shared memory environment OpenMP to analyze the relationship between parallel amount of calculation and the number of the processor core.The results of experiment,which were programmed by MPI to implement the row-divided algorithm and CANNON algorithm of matrix multiplication,point out that only in the comprehensive consideration of the architecture feature of multi-core system,system software,multi-core programming language environment and the correct use of algorithm can we design a better parallel application—being capable of high efficiency and small energy consumption.

Key words: Green computing,Memory alignment,OpenMP,CANNON algorithm,Multi-core processor

[1] http://wenku.baidu.com/view/e67cbf630b1c59eef8c7b457.html
[2] Kai Hwang.Advanced Computer Architecture:Parallelism Scalability Programmability[M].New York:McGraw-Hill Inc.,1993
[3] 王文义,董绍静.大规模并行处理系统及其程序设计方法研究——Cache缺失延迟、层次算法和可定域性[J].计算机研究与发展,1999,6(5):78-82 Wang Wen-yi,Dong Shao-jing.Large scale parallel processing system and its program design——Cache deletion delay,hierarchial algorithm and localizability[J].Journal of Computer Research and Development,1996,6(5):78-82
[4] 罗秋明,明仲,刘刚,等.OpenMP编译原理及实现技术[M].北京:清华大学出版社,2012:20-49
[5] 刘胜飞,张云泉,孙相征.一种改进的OpenMP指导调度策略研究[J].计算机研究与发展,2010,7(4):687-694 Liu Sheng-fei,Zhang Yun-quan,Sun Xiang-zheng.An Improved Guided Loop Scheduling Algorithm for OpenMP[J].Journal of Computer Research and Development,2010,7(4):687-694
[6] 徐磊,徐莹,张丹丹.多核构架下OpenMP多线程应用运行性能的研究[J].计算机工程与科学,2009,1(11):50-53 Xu Lei,Xu Ying,Zhang Dan-dan.A Study of the Open MP Multithread Application Execution Performance on Multicore Architecures[J].Computer Engineering and Science,2009,1(11):50-53
[7] Thoman P,Jordan H,Pellegrini S,et al.Automatic OpenMPloop scheduling:a combined compiler and runtime approach[C]∥Proceedings of 8th International Workshop on OpenMP.Rome,2012:88-101
[8] Shameem A,Jason R.多核程序设计技术-通过软件多线程提升性能[M].李宝峰,富弘毅,李韬,译.北京:电子工业出版社,2007:145-283 Shameem A,Jason R.Multi-core Programming:Increasing performance Through Software Multi-threading[M].Intel Corporation,2006
[9] 都志辉.高性能计算之并行编程技术—MPI并行程序设计[M].北京:清华大学出版社,2001:52-68
[10] Group W,Luck E,Skjellum A.Using MPI:Portable Parallel Programming with the Message Passing Interface[M].Cambridge,MA:MIT Press,1999
[11] Brown R.Performance and Productivity Comparison Between OpenMP and MPI[J].Int Parallel Prog,2007,5:441-458
[12] 剡公孝,申卫昌,刘骊,等.一种基于MPICH的高效矩阵相乘并行算法[J].计算机工程与应用,2009,5(26):72-73 Yan Gong-xiao,Shen wei-chang,Liu Li,et al.Effective matrix multiplication parallel algorithm based on MPICH[J].Computer Engineering and Applications,2009,5(26):72-73
[13] 王之元,胡庆丰,陈娟.能耗并行加速比:高性能计算系统综合性能的有效度量[J].计算机工程与科学,2009,1(11):113-116 Wang Zhi-yuan,Hu Qing-feng,Chen Juan.Power Parallel Speedup:An Effective Metric for Evaluating the Comprehensive Performance of High-Performance Computing Systems[J].Computer Engineering and Science,2009,1(11):113-116

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!