聚合通信模型的测试与分析  

Testing and Analyzing of Collective Communication Models

作　　者：刘洋[1,2] 曹建文[1,2] 李玉成[1,2] 

机构地区：[1]中科院软件所并行计算实验室,北京100080 [2]中国科学院研究生院,北京100039

出　　处：《计算机工程与应用》2006年第9期30-33,共4页Computer Engineering and Applications

摘　　要：聚合通信的性能影响着大规模并行数值计算软件的整体效率,它是基于点对点消息传递实现的。可以通过选择点对点通信模型来描述聚合通信。一个恰当的模型可以为聚合通信算法提供精确的定量分析,有助于算法的设计与研究。论文通过对若干聚合通信典型算法的测试与分析,尝试比较与定量分析两种点对点通信模型对聚合通信的描述精度。The performance of collective communications impacts the efficiency of large scale parallel numerical computing software.Collective communications are usually based on point-to-point communication.So we can use point- to-point models to characterize these collective communications.An accurate model can provide necessary evidences for quantitative analysis,which helps to design and analyze for algorithms.In this paper,we have tested and compared the precision of two models of collective communications on DeepComp 1800 by testing the three representative algorithms for reduction.

关 键 词：聚合通信 LogGP模型 Send/Recv 模型 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]