[1]董仁举 祝永志.并行编程模型的研究与发展[J].计算机技术与发展,2011,(01):92-94.
 DONG Ren-ju,ZHU Yong-zhi.Research and Development on Parallel-Programming Model[J].,2011,(01):92-94.
点击复制

并行编程模型的研究与发展()
分享到:

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2011年01期
页码:
92-94
栏目:
智能、算法、系统工程
出版日期:
1900-01-01

文章信息/Info

Title:
Research and Development on Parallel-Programming Model
文章编号:
1673-629X(2011)01-0092-03
作者:
董仁举 祝永志
曲阜师范大学计算机科学学院
Author(s):
DONG Ren-ju ZHU Yong-zhi
Department of Computer, Qufu Normal University
关键词:
并行编程消息传递共享变量两级并行模型TBB
Keywords:
parallel-programming message-passing shared memory two-level parallel model TBB
分类号:
TP311
文献标志码:
A
摘要:
并行编程模型在分布式计算中发挥着很重要的作用,随着人们对高性能计算需求的不断扩大和各种新技术的出现,并行编程模型也处于不断的发展和完善之中。对两种主要的编程模型进行了详细的分析和研究,针对前两种模型的优缺点分析并研究了两级并行模型的使用范围和优势等,最后针对硬件的新发展提出了新的编程模型的发展TBB+MPI。并在基于CMP的集群系统中实现了矩阵相乘的算法。实验结果显示TBB+MPI在多核集群编程方面有明显的优势,因此模型TBB+MPI更适合于多核集群。
Abstract:
Parallel-programming model takes a very important part in distributed computing. With the increasing need for high performance computing and the appearances of many new technologies,parallel-programming model is also in need of exploration and improvement. Two major parallel-programming models are compared in details in many aspects at first. Against to the advantages and disadvantages of the first two models, the usage and advantages of the two-level parallel is studied. According to the development of hardware, given the future trend of parallel-prograrmning model TBB+MPI. And has realized the matrix multiplication algorithm based on the CMP cluster system. The experiment result showed that the new model got the performance improved

相似文献/References:

[1]祝君 林庆农 徐造林.实时历史数据库中压缩技术的并行化研究[J].计算机技术与发展,2010,(07):36.
 ZHU Jun,LIN Qing-nong,XU Zao-lin.Research on Parallel Compression Technology in Real-Time Historical Database[J].,2010,(01):36.
[2]张翠莲 刘方爱 王亚楠.基于MPI的并行程序设计[J].计算机技术与发展,2006,(08):72.
 ZHANG Cui-lian,LIU Fang-ai,WANG Ya-nan.Parallel Program Design Based on MPI[J].,2006,(01):72.

备注/Memo

备注/Memo:
山东省高等学校实验研究项目基金(2005-400);曲阜师范大学校级科研项目(XJ0734)董仁举(1985-),男,硕士研究生,研究方向为分布式计算;祝永志,教授,硕士生导师,研究方向为网络与分布式系统
更新日期/Last Update: 1900-01-01