Abstract
Computational science problems with adaptive meshes involve dynamic load balancing when implemented on parallel machines. This dynamic load balancing requires fast partitioning of computational meshes at run time. We present in this report a scalable parallel partitioner, called S-HARP. The underlying principles of S-HARP are the fast feature of inertial partitioning and the quality feature of spectral partitioning. S-HARP partitions a graph from scratch, requiring no partition information from previous iterations. Two types of parallelism have been exploited in S-HARP, fine-grain loop-level parallelism and coarse-grain re cursive parallelism. The parallel partitioner has been implemented in Message Passing Interface on Cray T3E and IBM SP2 for portability. Experimental results indicate that S-HARP can partition a mesh of over 100,000 vertices into 256 partitions in 0.2 seconds on a 64-processor Cray T3E. S-HARP is much more scalable than other dynamic partitioners, giving over 15-fold speedup on 64 processors while ParaMeTiS 1.0 gives a few-fold speedup. Experimental results demonstrate that S-HARP is three to 10 times faster than the dynamic partitioners ParaMeTiS and Jostle on six computational meshes of size over 100,000 vertices.
1.This work is supported in part by the NASA JOVE Program, by travel support from USRA RIACS, and by summer support from MRJ, NASA Ames Research Center.
2.This work was supported by the Director, Office of Computational Sciences of the U.S. Department of Energy under contract number DE-AC03-76SF00098.
Preview
Unable to display preview. Download preview PDF.
References
S. T. Barnard and H. D. Simon, Fast multilevel implementation of recursive spectral bisection for partitioning unstructured problems, Concurrency: Practice and Experience 6, 1994, pp.101–117.
T. Chan, J. Gilbert, and S. Teng. Geometric spectral partitioning. Xerox PARC Technical Report, January 1995.
S. Guattery and G. L. Miller, On the performance of the spectral graph partitioning methods, in Proc. Sixth Annual ACM-SIAM Symposium on Discrete Algorithms, 1995, pp.233–242.
K. Hall, An r-dimensional quadratic placement algorithm, Management Science 17, November 1970, pp.219–229.
B. Hendrickson and R. Leland, A Multilevel Algorithm for Partitioning Graphs, in Proc. Supercomputing `95.
G. Karypis and V. Kumar, A fast and high quality multilevel scheme for partitioning irregular graphs, Tech. Report 95-035, University of Minnesota, 1995.
B. Nour-Omid, A. Raefsky and G. Lyzenga, Solving Finite Element Equations on Concurrent Computers, in Parallel Computations and their Impact on Mechanics, Ed. A.K. Noor, ASME, New York, 1986, p.209.
L. Oliker, Personal communication on the results of ParaMeTiS 1.0 and Jostle on SP2 and T3E, July 30, 1997.
K. Schloegel, G. Karypis, and V. Kumar, Parallel Multilevel Diffusion Schemes for Repartitioning of Adaptive Meshes, Tech. Report, Univ. of Minnesota, 1997.
H. D. Simon, Partitioning of unstructured problems for parallel processing, Computing Systems in Engineering, Vol. 2, 1991, pp. 135–148.
H. D. Simon, A. Sohn, and R. Biswas, HARP: A dynamic spectral partitioner, Journal ofParallel and Distributed Computing 50, April 1998, pp.88–103.
A. Sohn and H. D. Simon, JOVE: A dynamic load balancing framework for adaptive computations on distributed-memory multiprocessors, Technical Report, NJIT CIS 94-60, September 1994. (Also in Proc. of the ACMSymposium on Par ACM Symposium Algorithms and Architectures, June 1996 and IEEE SPDP, October 1996.)
A. Sohn and H. D. Simon, S-HARP: A parallel dynamic spectral partitioner, Technical Report, NJIT CIS 97-20, September 1997.
A. Sohn and R. Biswas, Special Issue on Dynamic Load Balancing, Journal of Parallel and Distributed Computing 47, December 1997, pp.99–101.
C. Walshaw, M. Cross, and M. Everett. Dynamic mesh partitioning: a unified optimization and load-balancing algorithm. Tech. Rep. 95/IM/06, University of Greenwich, London SE 18 6PF, UK, 1995.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sohn, A., Simone, H. (1998). S-HARP: A parallel dynamic spectral partitioner. In: Ferreira, A., Rolim, J., Simon, H., Teng, SH. (eds) Solving Irregularly Structured Problems in Parallel. IRREGULAR 1998. Lecture Notes in Computer Science, vol 1457. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0018554
Download citation
DOI: https://doi.org/10.1007/BFb0018554
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64809-3
Online ISBN: 978-3-540-68533-3
eBook Packages: Springer Book Archive