• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2012, Vol. 34 ›› Issue (8): 125-130.

• • 上一篇    下一篇

WCNS高精度并行软件的大规模计算研究

王运涛,王光学,徐庆新,邓小刚   

  1. (1.中国空气动力研究与发展中心空气动力学国家重点实验室,四川 绵阳621000;2.国防科学技术大学计算机学院,湖南长沙410073)

  • 收稿日期:2012-04-28 修回日期:2012-06-11 出版日期:2012-08-25 发布日期:2012-08-25
  • 基金资助:

    国家973计划资助项目(2009CB723801);空气动力学国家重点实验室基金资助项目资助项目(SKLA2009A0105)

A Study on Massively Parallel Computation#br#

WANG Guang-xue ,ZHANG Yu-lun,LI Song,DENG Xiao-gang   

  1. (1. State Key Laboratory of Aerodynamics,
    China Aerodynamics Research and Development Center, Mianyang 621000;
    2. School of Computer Science, National University of Defense Technology, Changsha 410073, China)

  • Received:2012-04-28 Revised:2012-06-11 Online:2012-08-25 Published:2012-08-25

摘要:

本文通过求解任意坐标系下的定常雷诺平均N-S方程和SST两方程湍流模型,采用五阶精度的加权紧致非线性格式(WCNS-E-5),实现流场的高精度数值模拟;基于分布式存储系统,采用MPI并行编程环境、非堵塞通信机制和遗传算法负载平衡,实现高精度模拟软件的并行化。在国防科学技术大学高性能计算应用研究中心的“天河”系统上完成软件移植、测试,通过对DLR-F6翼身组合体的模拟,说明软件并行策略和开发的正确性。最后,实现某民机全机的高精度并行模拟,网格规模达到1亿,为下一步WCNS高精度并行软件的大规模工程实际应用打下了坚实基础。

关键词: 加权紧致非线性格式, 大规模并行, 非堵塞, 数值模拟

Abstract:

Profiting from the fifth-order accurate explicit WCNS, high-order numerical simulations softwareis developed by solving the Reynolds Averaged Navier-Stokes equations together with the SST two-equation tur-bulence model in curvilinear coordinates. Based on the distributed storage system, a parallel version is accomplished adopting the MPI parallel programming environment, non-blocking algorithm and greedy-algorithmbased load balance strategy. Software is moved to and tested at the Galaxy stars system of the High PerformanceComputing Application Research Center of NUDT. Simulation of DLR-F6 shows that the parallel strategy isproper and the development is valid. Finally, high-order parallel simulation of some full civil aircraft is imple-mented, whose gird points number is 105 million. This work will lay a solid foundation for the future massiveengineering application of WCNS high-order parallel software.


Key words: weighted compact nonlinear schemes, massively parallel, non-block, numeric simulation

中图分类号: