visitors

Hybrid MPI/OpenMP parallelism for problems of hemodynamics
see also: zoom view aneurysm more
We present a hybrid parallelization framework to solve large scale problems of hemodynamics using OpenMP in conjunction with MPI environment. Nowadays, multi-core processors became a universal standard and are widely exploited at all levels of high-end workstations and scientific HPC clusters. We demonstrate that computational performance of a finite element code could be dramatically improved, when a proper use of a multi-core architecture is carried out. On a shared-memory system with two quad-core processors, a hybrid parallelization framework performs considerably faster than a native thread-based OpenMP implementation. We present a detailed roadmap to solve a 7.5 million degrees of freedom problem using an open source FreeFem++ platform, namely the MPI version. Three-dimensional models of an arterial bifurcation with either saccular or fusiform aneurysm have been constructed for benchmarking puproses.
References
[1] D. N. Arnold, F. Brezzi, and M. Fortin, A stable finite element for the Stokes equations, Calcolo, 21 (1984), pp. 337-344 (1985).
[2] J. M. Bull, J. Enright, X. Guo, C. Maynard, and F. Reid, Performance evaluation of mixed- mode OpenMP/MPI implementations, Int. J. Parallel Prog., 38 (2010), pp. 396-417.
[3] A. Buttari, J. Langou, J. Kurzak, and J. Dongarra, A class of parallel tiled linear algebra algorithms for multicore architectures, Parallel Comput., 35 (2009), pp. 38-53.
[4] B. Chapman, G. Jost, and R. van der Pas, Using OpenMP: Portable Shared Memory Parallel Programming, The MIT Press, 2010.
[5] T. A. Davis, Direct methods for sparse linear systems, vol. 2 of Fundamentals of Algorithms, Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA, 2006.
[6] M. Faverge and P. Ramet, A NUMA aware scheduler for a parallel sparse direct solver, Parallel Computing, (2009). Submitted.
[7] C. Geuzaine and J.-F. Remacle, Gmsh: A 3-D finite element mesh generator with built-in pre- and post-processing facilities, Internat. J. Numer. Methods Engrg., 79 (2009), pp. 1309-1331.
[8] R. Glowinski, Finite element methods for incompressible viscous flow, in Handbook of numerical analysis, Vol. IX, Handb. Numer. Anal., IX, North-Holland, Amsterdam, 2003, pp. 3-1176.
[9] F. Hecht, O. Pironneau, A. Le Hyaric, and K. Ohtsuka, FreeFem++, http://www.freefem.org.
[10] P. Henon, P. Ramet, and J. Roman, On finding approximate supernodes for an effcient block- ILU(k) factorization, Parallel Comput., 34 (2008), pp. 345-362.
[11] K. Johnson, P. Sharma, and J. Oshinskiab, Coronary artery flow measurement using navigator echo gated phase contrast magnetic resonance velocity mapping at 3.0T, Journal of Biomechanics, 41 (2008), pp. 595-602.
[12] W. E. Nagel, D. Kroner, and M. Resch, eds., High performance computing in science and engi- neering, Springer-Verlag, Berlin, 2008.
[13] W. W. Nichols and M. F. O'Rourke, Blood flow in arterie, Oxford University Press, 1998.
[14] O. Pironneau, Finite element methods for fluids, John Wiley & Sons Ltd., Chichester, 1989. Translated from the French.
[15] C. A. Rivera, M. Heniche, R. Glowinski, and P. A. Tanguy, Parallel finite element simulations of incompressible viscous fluid flow by domain decomposition with Lagrange multipliers , J. Comput. Phys., 229 (2010), pp. 5123-5143.
[16] L. R. Scott, T. Clark, and B. Bagheri, Scientific Parallel Computing, Princeton University Press, Princeton, NJ, USA, 2005.
[17] C. Taylor and P. Hood, A numerical solution of the Navier-Stokes equations using the finite element technique, Internat. J. Comput. & Fluids, 1 (1973), pp. 73-100.