作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2008, Vol. 34 ›› Issue (1): 100-101,. doi: 10.3969/j.issn.1000-3428.2008.01.033

• 软件技术与数据库 • 上一篇    下一篇

LBGK模型的分布式并行计算

段杰峰1,李 青1,周美莲2,蔡 群1   

  1. (1. 上海大学计算机工程与科学学院,上海 200072;2. 上海大学机电工程与自动化学院,上海200072)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-01-05 发布日期:2008-01-05

Distributed Parallel Computing for LBGK Model

DUAN Jie-feng1, LI Qing1, ZHOU Mei-lian2, CAI Qun1   

  1. (1. School of Computer Engineering and Science, Shanghai University, Shanghai 200072; 2. College of Mechatronics Engineering and Automation, Shanghai University, Shanghai 200072)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-01-05 Published:2008-01-05

摘要: 以二维9速模型D2Q9(LBGK)为例,针对不同集合形状的流场,研究了数据分布与通信量及并行效率之间的关系。在“自强3000”集群式超级计算机上就流场网格的不同规模、多种数据分布及通信方案进行了数值实验。测试结果表明,LBGK模型的并行效率高、可扩展性好,在数据分布方案与流场网格形状相似时,并行效率最高。该结论与理论分析相吻合。

关键词: 格子模型方法, 消息传递接口, 并行计算, 可扩展性

Abstract: This paper discusses the distributed parallel computing for LBGK D2Q9 model which is based on lattice Boltzmann method, and studies the influence of distribution of two-dimensional data and its communication operation on parallel efficiency. Lots of numerical experiments on "Ziqiang3000" supercomputers are done and experimental results show that LBGK model has good parallelism, scalability and parallel efficiency, which matches the theoretical analysis.

Key words: lattice model method, Message Passing Interface (MPI), parallel computing, scalability

中图分类号: