Optimizing Memory Constrained Environments in Monte Carlo Nuclear Reactor Simulations

Abstract

Monte Carlo neutron transport codes are a growing subject of research in nuclear reactor analysis. For robust reactor analysis, large scale neutron transport simulations require computation of reaction rates for tens of billions of particles involving several hundred isotopes. When employing physical-space domain decomposition, minimizing memory consumption while safely and efficiently exchanging massive amounts of data is a significant challenge. To address this problem, we implement and test several “memory-aware”, in-place, sparse, all-to-all MPI communication implementations. The algorithms are developed and tested within the open source MADRE (Memory-Aware Data Redistribution) project, which gives application programmers a simple API and set of tools and algorithms for carrying out memory-transparent in-place communication. We explore memory and communication efficiency tradeoffs for a range of in-place algorithms using a simple Monte Carlo communication kernel intended to mimic the behavior of our full Monte Carlo neutronics code.

Keywords

Monte Carlo neutron reactor analysis limited memory domain decomposition

Get full access to this article

View all access options for this article.

References

Chang

Cho

N. Z.

(2009). Some outstanding Problems in neutron transport computation. J. Korean Nuc. Soc. 41 (4), 381–390.

Hofmann

Rünger

(2009). Fine-grained data distribution operations for particle codes. In Proceedings of the 16th European PVM/MPI Users’ Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface, Berlin, Heidelberg, pp. 54–63. Springer-Verlag.

Hofmann

Rünger

(2010). An in-place algorithm for irregular all-to-all communication with limited memory. In Keller

Gabriel

Resch

Dongarra

(Eds.), Recent Advances in the Message Passing Interface, Volume 6305 of Lecture Notes in Computer Science, pp. 113–121. Springer Berlin / Heidelberg.

Hoogenboom

J. E.

Martin

W. R.

(2009). A proposal for a benchmark to monitor the performance of detailed Monte Carlo calculation of power densities in a full size reactor core. In Proc. 2009 Int. Conf. on Math., Comp. Meth., & Reactor Phys., Saratoga Springs, NY. American Nuclear Society.

Kelly

D. J.

Sutton

T. M.

Trumbull

Dobreff

P. S.

(2010). MC21 Monte Carlo analysis of the Hoogenboom Martin full-core PWR benchmark problem. In PHYSOR Advances in Reactor Physics to Power the Nuclear Renaissance.

Kumar

Dozsa

Almasi

Heidelberger

Chen

Giampapa

M. E.

Michael

Faraj

Parker

Ratterman

Smith

Archer

C. J.

(2008). The deep computing messaging framework: generalized scalable message passing on the blue gene/p supercomputer. In Proceedings of the 22nd annual international conference on Supercomputing, ICS ’08, New York, NY, USA, pp. 94–103. ACM.

Pinar

Hendrickson

(2004, July). Interprocessor communication with limited memory. IEEE Trans. Parallel Distrib. Syst. 15, 606–616.

Romano

Forget

Brown

(2010). Towards Scalable Parallelism in monte carlo particle transport codes using remote memory access. Progress in Nuclear Science and Technology 2, 670 –675.

Romano

P. K.

Forget

(2012). The OpenMC Monte Carlo Particle Transport Code. Ann. Nucl. Energy. Submitted.

10.

Siegel

Smith

Fischer

Mahadevan

(2011). Analysis of communication costs for domain decomposed monte carlo methods in nuclear reactor anlaysis. Journal of Computational Physics (0), –.

11.

Siegel

(2008). The MADRE Web Page. http://vsl.cis.udel.edu/madre.

12.

Siegel

S. F.

Siegel

A. R.

(2009). A memory-efficient data redistribution algorithm. In Ropo

Westerholm

Dongarra

(Eds.), Recent Advances in Parallel Virtual Machine and Message Passing Interface, 16th European PVM/MPI User’s Group Meeting, Proceedings, Volume 5759 of LNCS , pp. 219–229. Springer.

13.

Siegel

S. F.

Siegel

A. R.

(2010, February). MADRE: The Memory-Aware Data Redistribution Engine . International Journal of High Performance Computing Applications 24 (1), 93–104.