Abstract
We propose a simple and efficient load-balancing scheme for parallel molecular dynamics simulation on distributed memory machines. It decomposes spatial domain of particles into disjoint parts, each of which corresponds with a processor and dynamically changes its shape to keep about the same number of particles throughout the simulation. In contrast to other similar schemes, ours requires no long-distance inter-processor communications but only those among adjacent processors (thus little communication overheads), whereas it still guarantees fast reduction of load-imbalance among the processors. It owes these advantages mainly to the following features: (1) The sufficiently correct global load information is effectively obtained with step-wise propagation of appropriate information via nearest neighbor communication. (2) In addition to the global load-balancing, another load-balancing procedure is also invoked on each processor without global load information in order to suppress rapid increase or decrease of loads. Thus, informations from remote processors can provide reliable values even after a certain period of delay. Further, we discuss how to select loads to migrate among processors so that spatial locality of the processors may be preserved. Through preliminary evaluation on an uniprocessor workstation, we have shown the scheme has strong potential for large-scale parallel molecular dynamics simulation on distributed memory machines or workstation clusters.
Preview
Unable to display preview. Download preview PDF.
References
Stephen E. DeBolt and Peter A. Kollman. AMBERCUBE MD. J. of Comp. Chemistry, 1993.
K. Esselink, B. Smit, and P. A. J. Hilbers. Efficient parallel implementation of molecular dynamics on a toroidal network: Multi-particle potentials. J. of Computer Physics, 1993.
R. Giles and P. Tamayo. A Parallel Scalable Approach to Short-Range Molecular Dynamics on CM-5. In Scalable High Performance Computing Conference, page 240. IEEE, 1992.
David F. Hegarty and M. T. Kechadi. Topology Preserving Dynamic Load Balancing for Parallel Molecular Simulations. In IEEE Supercomputing. November 1997.
Jean-Marc Jézéquel and Jean-Lin Pacherie. Object-Oriented Application Frameworks, chapter EPEE: A Framework for Supercomputing. John Wiley & Sons, New York, 1998.
J. Kitowski. Distributed and parallel computing of short-range molecular dynamics. Lecture Notes in Computer Science, 1041, 1996.
Kian-Tat Lim. Molecular Dynamics for Very Large Systems on Massively Parallel Computers: The MPSim Program. J. of Comp. Chemistry, 1997.
S. L. Lin, J. Mellor-Crummey, B. M. Pettitt, and G. N. Phillips Jr. Molecular Dynamics on a Distributed-Memory Multiprocessor. J. of Comp. Chemistry, 1992.
Steve Plimpton. Fast Parallel Algorithms for Short-Range Molecular Dynamics. J. of Comp. Physics, 1995.
Naohito Sato, Satoshi Matsuoka, Jean-Marc Jézéquel, and Akinori Yonezawa. A Methodology for Specifying Data Distribution using only Standard Object-Oriented Features. In the 11th International Conference on Supercomputing. ACM SIGARCH, July 1997.
P. K. Weiner and P. A Kollman. AMBER. J. of Comp. Chemistry, 1981.
Marc H. Wellebeek-LeMair and Anthony P. Reeves. Strategies for Dynamic Load Balancing on Highly Parallel Computers. IEEE Trans. on Parallel and Distributed Systems, 1993.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1999 Springer-Verlag
About this paper
Cite this paper
Sato, N., Jézéquel, JM. (1999). A simple dynamic load-balancing scheme for parallel molecular dynamics simulation on distributed memory machines. In: Sloot, P., Bubak, M., Hoekstra, A., Hertzberger, B. (eds) High-Performance Computing and Networking. HPCN-Europe 1999. Lecture Notes in Computer Science, vol 1593. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0100587
Download citation
DOI: https://doi.org/10.1007/BFb0100587
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65821-4
Online ISBN: 978-3-540-48933-7
eBook Packages: Springer Book Archive