| 基于微机集群系统的MPI并行计算 |
| |
| 引用本文： | 卢云娥,黄宗宇,李超阳,郭祥斌,殷慧明. 基于微机集群系统的MPI并行计算[J]. 电子设计工程, 2011, 19(5): 78-81 |
| |
| 作者姓名： | 卢云娥 黄宗宇 李超阳 郭祥斌 殷慧明 |
| |
| 作者单位： | 长沙理工大学计算机与通信工程学院,湖南,长沙,410114 |
| |
| 基金项目： | 教育部2010年国家级大学生创新性实验计划资助项目 |
| |
| 摘 要： | 计算机集群系统具有性能高、扩展性强等特点,能够充分发挥并行计算能力。在现有的微机条件上建立了基于Windows和MPI的集群计算环境,并在此平台上通过MPI并行程序设计思想,提出了利用蒙特卡罗积分法的PI值求解算法。通过对大量的PI值计算实验结果分析,发现基于MPI的并行计算能大幅度提高运算效率。
|
| 关 键 词： | 集群 并行计算 MPI PI |
The MPI parallel computing based on the microcomputer cluster |
| |
| LU Yun-e,HUANG Zong-yu,LI Chao-yang,GUO Xiang-bin,YIN Hui-ming. The MPI parallel computing based on the microcomputer cluster[J]. Electronic Design Engineering, 2011, 19(5): 78-81 |
| |
| Authors: | LU Yun-e HUANG Zong-yu LI Chao-yang GUO Xiang-bin YIN Hui-ming |
| |
| Affiliation: | (Department of Computer and Communication Engineering,Changsha University of Science and Technology,Changsha 410114,China) |
| |
| Abstract: | Computer cluster system has some characteristics such as the high performance,scalable and so on,which can make full use of parallel computation ability.And it has built computing environment which is based on the Windows and MPI cluster in the existing conditions of the microcomputer,and on this platform through MPI parallel programming ideas,presented using the Monte Carlo integration method of PI values algorithm.Based on large amounts of PI value calculation analysis of experimental results,found that based on the MPI parallel computation can dramatically improve operation efficiency. |
| |
| Keywords: | cluster system parallel computing message passing inerface PI |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
