VecDualSPHysics: A vectorized implementation of Smoothed Particle Hydrodynamics method for simulating fluid flows on multi-core processors

VecDualSPHysics: A vectorized implementation of Smoothed Particle Hydrodynamics method for simulating fluid flows on multi-core processors
复制标题

VecDualSPHysics:平滑粒子流体动力学方法的矢量化实现,用于模拟多核处理器上的流体流动

DOI:
10.1016/j.jcp.2022.111234
复制
发表时间:
2022-04
影响因子:
4.1
通讯作者:
Canqun Yang
Canqun Yang
中科院分区:
物理与天体物理2区
文献类型:
--
作者:
Sifan Long;Xiaokang Fan;Chao Li;Yi Liu;Sijiang Fan;Xiao-Wei Guo;Canqun Yang

文献摘要

参考文献

被引文献

相似文献

作为最有前途的无网格数值方法之一,光滑粒子流体力学(SPH)算法可以用来模拟许多工程问题。然而,要实现超大规模的并行仿真,就必须充分利用计算处理器的性能。在本文中,我们专注于在现代多核处理器上通过向量化来最大化SPH程序的性能。总结了向量化的一般原理,包括分支向量化、向量约简和指令集优化方法。然后,基于这些基本矢量化原则,实现了矢量化版本的VecDualSPH。实验表明,VecDualSP物理在性能上比原来的DualSP物理有了很大的提高。通过将AVX2指令集与OpenMP相结合,在单核多核处理器上获得了4.3倍的加速比,AVX-512指令集的最大加速比为6.3倍,远远快于纯基于OpenMP的多线程版本。VecDualSP物理显示了通过矢量化来开发多核处理器性能的巨大潜力,并可进一步用于优化大规模并行SPH模拟。
As one of the most promising mesh-free numerical approaches, the Smoothed Particle Hydrodynamics (SPH) algorithm can be used for simulating many engineering problems. However, it is necessary to fully exploit the performance of computing processors for realizing extremely large-scale parallel simulations. In this paper, we focused on maximizing the performance of a SPH program by vectorizing on modern multi-core processors. We summarize the general principles of vectorization including the branch vectorization, vector reduction and instruction set optimization approaches. Then, a vectorized version VecDualSPHysics was implemented based on these basic vectorization principles. Experiments show that VecDualSPHysics has a significant performance improvement over the original DualSPHysics. By using AVX2 instruction set combined with OpenMP, we obtained up to 4.3x speedups on a single multi-core processor, and the maximal speedup is 6.3x for AVX-512 instruction set, which is much faster than the pure OpenMP-based multi-threading version. The VecDualSPHysics shows the great potential of exploiting the performance of multi-core processors by vectorization and can further be used to optimize the massively parallel SPH simulations.
SPH 的新核函数及其在自由表面流中的应用
DOI: 10.1016/j.apm.2013.12.001
发表时间: 2014-08
影响因子: 5
作者:
X.F. Yang;S.L. Peng;M.B. Liu
通讯作者: M.B. Liu
DOI: 10.1016/j.crma.2018.12.002
发表时间: 2018-05
影响因子: 0.8
作者:
Paul Acevedo;C. Amrouche;C. Conca;Amrita Ghosh
通讯作者: Paul Acevedo;C. Amrouche;C. Conca;Amrita Ghosh
DOI: 10.1145/2614348.2614375
发表时间: 2007-04
期刊: --
影响因子: --
作者:
T. Harada;S. Koshizuka;Y. Kawaguchi
通讯作者: T. Harada;S. Koshizuka;Y. Kawaguchi
DOI: 10.4103/0970-9290.70813
发表时间: 2010-07
期刊: Indian journal of dental research : official publication of Indian Society for Dental Research
影响因子: --
作者:
A. Srirekha;Kusum Bashetty
通讯作者: A. Srirekha;Kusum Bashetty
DOI: 10.1016/j.cpc.2017.11.015
发表时间: 2017-10
期刊: ArXiv
影响因子: --
作者:
H. Homann;Francois Laenen
通讯作者: H. Homann;Francois Laenen