基于OpenMP矩阵相乘并行算法的设计  

A research on parallel algorithm of matrix multiplication based on OpenMP

作　　者：潘亮[1] 郭改枝[1] 宋鑫梦 

机构地区：[1]内蒙古师范大学计算机与信息工程学院,内蒙古呼和浩特010022

出　　处：《宝鸡文理学院学报（自然科学版）》2014年第1期21-23,共3页Journal of Baoji University of Arts and Sciences(Natural Science Edition)

基　　金：内蒙古自然科学基金资助项目(2010MS0918)

摘　　要：目的设计并实现一种基于数据划分的矩阵乘法的并行算法,将划分的数据交给多个线程同时执行,充分挖掘计算机的性能。方法根据OpenMP并行编程的基本风格,并在Visual Studio2005上搭建能够实现并行编程的环境平台。结果并行算法所花费的时间较非并行算法短。结论通过与非并行矩阵乘法性能进行比较,验证该算法可以有效地利用多核处理器的优势。Objective-To design and implement the parallel algorithm of matrix multiplication based on data partitioning so that the partitioned data are provided for multiple threads to execute con currently and the computer＇s performance is fully tapped. Methods-According to the basic style of parallel programming based on OpenMP, a parallel programming environment platform is established on the Visual Studio 2005. Results-The time spent in the parallel algorithm is shorter than that in non-parallel algorithms. Conclusion-Compared with the performance of non-parallel matrix multipli cation, it is proved that the algorithm can effectively utilize the advantages of multi-core processors.

关 键 词：数据划分 矩阵乘法 并行算法 OPENMP 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]