AUTHOR=Long Sifan , Wong Kelvin K. L. , Fan Xiaokang , Guo Xiaowei , Yang Canqun 

TITLE=Smoothed particle hydrodynamics method for free surface flow based on MPI parallel computing

JOURNAL=Frontiers in Physics

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/physics/articles/10.3389/fphy.2023.1141972

DOI=10.3389/fphy.2023.1141972

ISSN=2296-424X

ABSTRACT=In the field of computational fluid dynamics, smoothed particle hydrodynamics (SPH), as a classical particle method, is very suitable for simulating large deformation, free surface flow and other types of flow scenarios. However, traditional SPH methods are inefficient to compute, resulting in application scenarios that are constrained by accuracy requirements. At present, DualSPHysics is an excellent SPH application software tool standard in academia. Harnessing this tool, our paper presents a large-scale parallel framework: parallelDualSPHysics, which can solve the simulation of large-scale fluid flow. First, an efficient domain decomposition algorithm is proposed, and the data structure of DualSPHysics in a parallel framework is reshaped, with different particles evenly distributed to different processors and collaborative computing. Secondly, a parallel particle interaction and particle update module is introduced, which enables the SPH solver to synchronize computations in multiple processes. Finally, the pre-processing and post-processing capabilities of the parallelDualSPHyiscs version enable the software to simulate large-scale engineering cases on a cluster of distributed memory. In addition, scalability test analysis shows that the 3 to 120 million particles tested in this paper can still maintain more than 90$\%$ computing efficiency, which demonstrates that the parallel strategy given in this paper can achieve very high parallel efficiency.