Rendezvous algorithms for large-scale modeling and simulation

Plimpton, Steven J.; Knight, Christopher

doi:10.1016/j.jpdc.2020.09.001

Rendezvous algorithms for large-scale modeling and simulation

Journal Article · Fri Sep 18 00:00:00 UTC 2020 · Journal of Parallel and Distributed Computing

DOI:https://doi.org/10.1016/j.jpdc.2020.09.001· OSTI ID:1668692

Plimpton, Steven J. ^[1]; ^[2]

Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Argonne National Lab. (ANL), Lemont, IL (United States)

Rendezvous algorithms encode a communication pattern that is useful when processors sending data do not know who the receiving processors should be, or vice versa. The idea is to define an intermediate decomposition where datums from different sending processors can ”rendezvous” to perform a computation, in a manner that both the senders and eventual receivers of the results can identify the appropriate rendezvous processor. Though they were originally designed for interpolating between overlaid grids with independent parallel decompositions (Plimpton et al., 2004), we have recently found rendezvous algorithms useful for a variety of operations in particle- or grid-based simulation codes when running large problems on large numbers of processors. In particular, we show they can perform well when a load-balanced intermediate decomposition is randomized and not spatial, requiring all-to-all communication to move data between processors. In this case rendezvous algorithms leverage the large bisection communication bandwidths which parallel machines provide. We describe how rendezvous algorithms work in a scientific computing context and give specific examples for molecular dynamics and Direct Simulation Monte Carlo codes which result in dramatic performance improvements versus simpler algorithms which do not scale as well. We explain how a generic rendezvous algorithm can be implemented, and also point out similarities with the MapReduce paradigm popularized by Google and Hadoop.

Research Organization:: Argonne National Laboratory (ANL), Argonne, IL (United States); Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)

Sponsoring Organization:: USDOE; USDOE National Nuclear Security Administration (NNSA)

Grant/Contract Number:: AC02-06CH11357; AC04-94AL85000; NA0003525

OSTI ID:: 1668692

Alternate ID(s):: OSTI ID: 2325459
OSTI ID: 1776628

Report Number(s):: SAND--2020-4542J; 685787

Journal Information:: Journal of Parallel and Distributed Computing, Journal Name: Journal of Parallel and Distributed Computing Vol. 147; ISSN 0743-7315

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (12)

Fast Parallel Algorithms for Short-Range Molecular Dynamics Plimpton, Steve Journal of Computational Physics, Vol. 117, Issue 1 https://doi.org/10.1006/jcph.1995.1039	journal	March 1995
Aspherical particle models for molecular dynamics simulation Nguyen, Trung Dac; Plimpton, Steven J. Computer Physics Communications, Vol. 243 https://doi.org/10.1016/j.cpc.2019.05.010	journal	October 2019
A framework approach for developing parallel adaptive multiphysics applications Stewart, James R.; Edwards, H. Carter Finite Elements in Analysis and Design, Vol. 40, Issue 12 https://doi.org/10.1016/j.finel.2003.10.006	journal	July 2004
Mesh-free data transfer algorithms for partitioned multiphysics problems: Conservation, accuracy, and parallelism Slattery, Stuart R. Journal of Computational Physics, Vol. 307 https://doi.org/10.1016/j.jcp.2015.11.055	journal	February 2016
A parallel rendezvous algorithm for interpolation between multiple grids Plimpton, Steven J.; Hendrickson, Bruce; Stewart, James R. Journal of Parallel and Distributed Computing, Vol. 64, Issue 2 https://doi.org/10.1016/j.jpdc.2003.11.006	journal	February 2004
MapReduce in MPI for Large-scale graph algorithms Plimpton, Steven J.; Devine, Karen D. Parallel Computing, Vol. 37, Issue 9 https://doi.org/10.1016/j.parco.2011.02.004	journal	September 2011
Probing the limits of metal plasticity with molecular dynamics simulations Zepeda-Ruiz, Luis A.; Stukowski, Alexander; Oppelstrup, Tomas Nature, Vol. 550, Issue 7677 https://doi.org/10.1038/nature23472	journal	September 2017
Direct simulation Monte Carlo on petaflop supercomputers and beyond Plimpton, S. J.; Moore, S. G.; Borner, A. Physics of Fluids, Vol. 31, Issue 8 https://doi.org/10.1063/1.5108534	journal	August 2019
A Partitioning Strategy for Nonuniform Problems on Multiprocessors No authors listed IEEE Transactions on Computers, Vol. C-36, Issue 5 https://doi.org/10.1109/TC.1987.1676942	journal	May 1987
An overview of the Trilinos project Heroux, Michael A.; Phipps, Eric T.; Salinger, Andrew G. ACM Transactions on Mathematical Software, Vol. 31, Issue 3 https://doi.org/10.1145/1089014.1089021	journal	September 2005
MapReduce: simplified data processing on large clusters Dean, Jeffrey; Ghemawat, Sanjay; Mehta, Brijesh Communications of the ACM, Vol. 51, Issue 1 https://doi.org/10.1145/1327452.1327492	journal	January 2008
Marching cubes: A high resolution 3D surface construction algorithm Lorensen, William E.; Cline, Harvey E. ACM SIGGRAPH Computer Graphics, Vol. 21, Issue 4 https://doi.org/10.1145/37402.37422	journal	August 1987

Similar Records

Performance modeling of the Ada rendezvous. Final report

Technical Report · Tue Oct 01 04:00:00 UTC 1991 · OSTI ID:5745722

Complexity theory for unbounded fan-in parallelism

Conference · Fri Jan 01 04:00:00 UTC 1982 · OSTI ID:5248275

Related Subjects

97 MATHEMATICS AND COMPUTING
MapReduce
Modeling and simulation
Parallel communication
Rendezvous algorithms
modeling and simulation
parallel communication

Rendezvous algorithms for large-scale modeling and simulation

Citation Formats

References (12)

Similar Records

Related Subjects