引用本文: | 刘枫,李桦,田正雨,等.基于MPI+CUDA的异构并行可压缩流求解器.[J].国防科技大学学报,2014,36(1):6-10.[点击复制] |
LIU Feng,LI Hua,TIAN Zhengyu,et al.Heterogeneous parallel compressible flow solver based on MPI+CUDA[J].Journal of National University of Defense Technology,2014,36(1):6-10[点击复制] |
|
|
|
本文已被:浏览 14763次 下载 7724次 |
基于MPI+CUDA的异构并行可压缩流求解器 |
刘枫1,2, 李桦1, 田正雨1, 潘沙1 |
(1.国防科技大学 航天科学与工程学院,湖南 长沙 410073;2.
2.中国空气动力研究与发展中心,四川 绵阳 621000)
|
摘要: |
在CPU/GPU异构体系结构计算集群上,建立了基于MPI+CUDA的异构并行可压缩流求解器。讨论了异构结构上的可压缩流并行算法的并行模式,在CPU上执行计算密集度低、指令复杂的计算任务,在GPU上执行计算密集度高、指令单一的计算任务。通过数个算例,对比了异构并行计算和传统CPU并行计算计算结果和计算效率。将该算法运用于高超声速流动的数值模拟中,数值结果显示,基于MPI+CUDA的异构并行可压缩流求解器鲁棒性好,计算效率较CPU同构并行计算提高10倍以上。 |
关键词: 消息传递接口 统一计算设备架构 异构计算 可压缩流 |
DOI:10.11887/j.cn.201401002 |
投稿日期:2013-05-09 |
基金项目:国家自然科学基金资助项目(91016010,91216117) |
|
Heterogeneous parallel compressible flow solver based on MPI+CUDA |
LIU Feng1,2, LI Hua1, TIAN Zhengyu1, PAN Sha1 |
(1.College of Aerospace Science and Engineering, National University of Defense Technology, Changsha 410073, China;2.
2.Chinese Aerodynamic Research and Development Center,Mianyang 621000,China)
|
Abstract: |
A compressible flow heterogeneous parallel solver based on MPI+CUDA on CPU/GPU heterogeneous system was established. Then different parallel computing models and optimizing methods of compressible flow parallel computing algorithm were discussed. This solver runs different codes with difference resources: the codes which are complex or have low computing density are run on CPU, while the codes which are simple or have high computing density are run on GPU. The heterogeneous systems’ computing results and the efficiencies with homogeneous systems were compared through several problems. Finally, the heterogeneous algorithm was applied to the hypersonic flow. The result shows that the algorithm is robust and the computing efficiency is improved ten times more than that of the homogeneous algorithm. |
Keywords: MPI(Message Passing Interface) CUDA(Compute Unified Device Architecture) heterogeneous computing compressible flow |
|
|
|
|
|