Hybrid MPI + OpenMP Parallelization of Scramjet Simulation with Hypergraph Partitioning

Abstract:

As one of the most significant methods to study hypersonic flight vehicle, the numerical simulation of supersonic combustion ramjet has drawn an ever increasing attention at present. Nevertheless, the traditional serial simulation cannot satisfy the practical needs because of high calculation precision, avaricious memory overhead and overlong computation time. In this paper, we study on a general algorithm for scramjet simulation, and bring about parallelization by using a hypergraph partitioning algorithm and a two level hybrid parallel model. The whole computing domain is decomposed into several sub-domains based on hypergraph partitioning, and each sub-domain is assigned to a MPI process. A single step of computation operates in the inter loop level, where a compiler directive is used to split MPI process into several OpenMP threads. Finally, speedup and parallel efficiency of our hybrid program testing on a China-made supercomputer with 4 to 256 cores is compared with pure MPI program. And, the hybrid program exhibits better parallel performance than the pure MPI program in the main, roughly as expected. The result indicates that our hybrid parallel strategy is effective and practical in scramjet simulation.

• Periodical: Advanced Materials Research (Volumes 712-715)
• Pages: 1294-1297
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.712-715.1294
• Online since: June 2013
• Authors: Yao Yuan Zeng, Wen Tao Zhao, Zheng Hua Wang
• Keywords: Hybrid Parallel Strategy, Hypergraph Partitioning, MPI, OpenMP, Parallel Performance, Scramjet

References

[1] Wilcox, D.C., Reassessment of the Scale-Determining Equation for Advanced Turbulence Models. AIAA Journal, 1988. 26(11): pp.1299-1310.

DOI: 10.2514/3.10041

[2] Menter, F.R., Two-Equation Eddy-Viscosity Turbulence Models for Engineering Applications. AIA A JOURNAL, 1994. 32(8): pp.1598-1605.

DOI: 10.2514/3.12149

[3] Kalitzin, G., A.R.B. Gould and J.J. Benton, Application of two-equation turbulence models in aircraft design, in 34th Aerospace Sciences Meeting and Exhibit. 1996: Reno, NV.

DOI: 10.2514/6.1996-327

[4] Spalart, P.R. and S.R. Allmaras, A ONE-EQUATION TURBULENCE MODEL FOR AERODYNAMIC FLOWS, in 30th Aerospace Sciences Meeting & ExhlbH. 1992: Reno, NV.

DOI: 10.2514/6.1992-439

[5] Trifunović, A. and W.J. Knottenbelt, Parallel multilevel algorithms for hypergraph partitioning. Journal of Parallel and Distributed Computing, 2008. 68(5): pp.563-581.

DOI: 10.1016/j.jpdc.2007.11.002

[6] Catalyurek, U.V., et al., Dynamic Load Balancing for Adaptive Scientific Computations via Hypergraph Repartitioning. 2009.

[7] Bansal, N., et al. Min-max graph partitioning and small set expansion. in Foundations of Computer Science (FOCS), 2011 IEEE 52nd Annual Symposium on. 2011: IEEE.

DOI: 10.1109/focs.2011.79

[8] Conan-Guez, B. and F. Rossi, Dissimilarity Clustering by Hierarchical Multi-Level Refinement. arXiv preprint arXiv:1204.6509, 2012.

[9] Rital, S., Hypergraph cuts & unsupervised representation for image segmentation. fundamenta informaticae, 2009. 96(1): pp.153-179.

DOI: 10.3233/fi-2009-172

[10] Mininni, P.D., et al., A hybrid MPI–OpenMP scheme for scalable parallel pseudospectral computations for fluid turbulence. Parallel Computing, 2011. 37(6): pp.316-326.

DOI: 10.1016/j.parco.2011.05.004