Abstract
A new algorithm of two level parallelization for direct simulation Monte Carlo is elaborated for solving unsteady problems of molecular gasdynamics on shared and hybrid memory multiprocessor computers. The first parallelization level (parallel statistically independent runs) is implemented with the aid of MPI library. For the second level (data parallelization) the standard UNIX interprocess communications are employed. Two versions of static load balancing are used for the second level. The study on speedup is carried out by solving one typical test problem of molecular gasdynamics. The computation is performed on shared memory HP/Convex SPP-1600 (8 processors) and hybrid memory PARITET (4 two processor nodes).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Ivanov M., Markelov G., Taylor S., Watts J. Parallel DSMC strategies for 3D computations. Proc. Parallel CFD’96. North Holland, Amserdam, 1997, pp.485–492.
Wilmoth R.G. Adaptive Domain Decomposition for Monte Carlo Simulations on Parallel Processor. In. 17th Rarefied Gas Dynamics Symposium. AIAA, July 1990.
A.V. Bogdanov, N.Y. Bykov, G.A. Lukianov. Distributed and Parallel Direct Simulation Monte Carlo of Rarefied Gas Flows. Lecture Notes in Computer Science, Vol. 1401. Springer-Verlag, Berlin Heidelberg New York (1998)
A.V. Bogdanov, I.A. Grishin, Gr.O. Khanlarov, G.A. Lukianov and V.V. Zakharov. Algorithm of Two-Level Parallelization for Direct Simulation Monte Carlo of Unsteady Flows in Molecular Gasdynamics. Lecture Notes in Computer Science, Vol.1593. Springer-Verlag, Berlin Heildelberg New York (1999).
G.A. Bird. Molecular Gasdynamics and Direct Simulation of Gas Flows. Clarendon Press. Oxford. 1994
P. Kuonen, R. Gruber, A. De Vita, P. Volgers. Parallel computer architectures for commodity computing. HPCN Europe 99. Late Papers. pp. 1–10.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Khanlarov, G.O., Lukianov, G.A., Malashonok, D.Y., Zakharov, V.V. (2000). Parallel DSMC on Shared and Hybrid Memory Multiprocessor Computers. In: Bubak, M., Afsarmanesh, H., Hertzberger, B., Williams, R. (eds) High Performance Computing and Networking. HPCN-Europe 2000. Lecture Notes in Computer Science, vol 1823. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45492-6_68
Download citation
DOI: https://doi.org/10.1007/3-540-45492-6_68
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-67553-2
Online ISBN: 978-3-540-45492-2
eBook Packages: Springer Book Archive