The Impact of the Mesh Partitioning Factors on CFD Simulation
Mesh partitioning plays an important role in Computational Fluid Dynamics (CFD) simulation. However, it is difficult to produce a good mesh partitioning to achieve a high performance simulation because it is a NP-complete problem to solve this problem. A good mesh partitioning may be determined by many factors. Nowadays, there is a lack of systematic guidelines and comprehensive analyses for how to produce a high quality mesh partitioning. Considering that it is difficult to break through in theory, we prefer to explore how the factors related to mesh partitioning influence CFD simulation based on a large amount of experimental analyses. In this paper, we evaluate the impact of changing the mesh partitioning factors on mesh partitioning quality and simulation performance according to the following five factors: the number of processor faces, subdomain aspect ratio, load, partitioning direction and mapping. We design a series of rules to make various and numerous mesh partitioning changes based on four commonly used methods (Simple, Metis, Scotch and Manual) in OpenFOAM. Furthermore, we conduct the parallel simulation for two representative cases (cavity case and damBreak case) in OpenFOAM framework. The experimental results certify that changing mesh partitioning factors really influences the simulation performance. Then we provide some analyses and advices, which will definitely be helpful to guide mesh partitioning in the future.
KeywordsMesh partitioning CFD simulation Mesh partitioning quality Simulation performance
The authors would like to thank to the funding from the National Natural Science Foundation of China under grant No.61221491, No.61303071, No.61303068, No.61303063, No.61303061 and No.61120106005, and Open Fund (No. 201402-01, No.201503-01 and No.201503-02) from State Key Laboratory of High Performance Computing.
- 1.Schloegel, K., Karypis, G., Kumar, V.: Graph partitioning for high performance scientific simulations. Army High Performance Computing Research Center (2000)Google Scholar
- 4.Aftosmis, M., Berger, M., Murman, S.: Applications of space-filling-curves to cartesian methods for CFD. In: 42nd AIAA Aerospace Sciences Meeting and Exhibit, p. 1232 (2004)Google Scholar
- 8.Vanderstraeten, D., Keunings, R.: Beyond conventional mesh partitioning algorithms and the minimum edge cut criterion: Impact on realistic applications. Technical report. Society for Industrial and Applied Mathematics, Philadelphia, PA (United States) (1995)Google Scholar
- 11.Bhatelé, A., Kalé, L.V., Kumar, S.: Dynamic topology aware load balancing algorithms for molecular dynamics applications. In: Proceedings of the 23rd International Conference on Supercomputing, pp. 110–116. ACM (2009)Google Scholar
- 12.Aktulga, H.M., Yang, C., Ng, E.G., Maris, P., Vary, J.P.: Topology-aware mappings for large-scale eigenvalue problems. In: Kaklamanis, C., Papatheodorou, T., Spirakis, P.G. (eds.) Euro-Par 2012. LNCS, vol. 7484, pp. 830–842. Springer, Heidelberg (2012). doi: 10.1007/978-3-642-32820-6_82 CrossRefGoogle Scholar
- 13.Karypis, G., Kumar, V.: A software package for partitioning unstructured graphs, partitioning meshes, and computing fill-reducing orderings of sparse matrices. Department of Computer Science and Engineering, Army HPC Research Center, University of Minnesota, Minneapolis, MN (1998)Google Scholar
- 14.Pellegrini, F., Roman, J.: Scotch: a software package for static mapping by dual recursive bipartitioning of process and architecture graphs. In: Liddell, H., Colbrook, A., Hertzberger, B., Sloot, P. (eds.) HPCN-Europe 1996. LNCS, vol. 1067, pp. 493–498. Springer, Heidelberg (1996). doi: 10.1007/3-540-61142-8_588 CrossRefGoogle Scholar
- 15.Walshaw, C., Cross, M.: JOSTLE: parallel multilevel graph-partitioning software – an overview. In: Magoules, F. (ed.) Mesh Partitioning Techniques and Domain Decomposition Techniques, pp. 27–58. Civil-Comp Ltd. (2007)Google Scholar
- 19.Wang, M., Tang, Y., Guo, X., Ren, X.: Performance analysis of the graph-partitioning algorithms used in OpenFOAM. In: 2012 IEEE Fifth International Conference on Advanced Computational Intelligence (ICACI), pp. 99–104. IEEE (2012)Google Scholar
- 22.OpenFOAM User Guide: Version 2.3.0, 5 February 2014Google Scholar