摘要
本文基于魔方(曙光5000A)"超并行"系统架构比较了不同并行编程模型下NAS Parallel Benchmark Multi-Zone的性能。结果表明,在单节点共享内存存储架构下OpenMP编程模型能获得更好性能,跨节点采用混合编程模型能获得更好性能,测试结果为百万亿次系统的使用及应用性能的提升起到指导作用。
The paper compared the NAS Parallel Benchmark Multi-Zone performance of different parallel programming model based on Magic Cube（Dawning 5000A） HPP architecture. The results show that OpenMP has a better performance on single SMP node, and hybrid parallel programming model offers better performance inter-nodes, which will provide guidance for the using a teraflops supercomputer system and improving the performance of applications.
出处
《计算机工程与科学》
CSCD
北大核心
2009年第A01期245-248,共4页
Computer Engineering & Science
基金
国家自然科学基金资助项目(90612017)
国家863计划资助项目(2006AA01A112)
国家863计划资助项目(2006AA01A117)
