Stephen T. Barnard
ABSTRACT
The design of a parallel implementation of multilevel recursive spectral bisection on the Cray T3D is described. The code is intended to be fast enough to enable dynamic repartitioning of adaptive meshes and to partition meshes that are too large for workstations. Two innovations in the implementation are recursive asynchronous task teams and a parallel version of the multilevel accelerator. A performance improvement of a factor of 140 over the best available serial implementation is demonstrated.
- 1.S. T. Barnard and H. D. Simon, 'Fast multilevel implementation of recursive spectral bisection for partitioning unstructured problems', Concurrency: Practice and Experience, Vol. 6, No. 2, 101-117 (1994).Google Scholar
Cross Ref
- 2.S. T. Barnard, A. Pothen, and H. D. Simon, 'A spectral algorithm for envelope reduction of sparse matrices', Proceedings of SU- PERCOMPUTING '93, 493-502, Portland, Oregon, (November 15-19, 1993). Google ScholarDigital Library
- 3.E. D. Brooks III, B. C. Gorda, and Karen H. Warren, 'The Parallel C Preprocessor', Scientific Programming, Vol. 1, No. 1, 79-89 (1992).Google ScholarDigital Library
- 4.CMSSL For CM Fortran, Vol. II, version 3.2, Thinking Machines Corp., (1994).Google Scholar
- 5.P. Diniz, S. Plimpton, B. Hendrickson, and R. Leland, 'Parallel algorithms for dynamically partitioning unstructured grids', Proceeding of the Seventh SIAM Conference on Parallel Processing for Scientific Computing, 615-620 (1995).Google Scholar
- 6.M. Garey, D. Johnson, and L. Stockmeyer, 'Some simplified NP- complete graph problems', Theoretical Computer Science,1, 237-267 (1976).Google ScholarCross Ref
- 7.C. Lanczos, 'An iteration method for the solution of the eigenvalue problem of linear differential and integral operators', J. Res. Nat. Bur. Stand., 45, 255-282 (1950).Google ScholarCross Ref
- 8.M. Luby, 'A simple parallel algorithm for the maximal independent set problem', SIAM J. Comput., Vol. 15, No. 4, (1986). Google ScholarDigital Library
- 9.C. C. Paige and M. A. Saunders, 'Solution of sparse indefinite systems of linear equations', SIAM J. Numer. Anal., Vol. 12, 617-629 (1974).Google ScholarDigital Library
- 10.B. N. Parlett, 'The Rayleigh quotient iteration and some generalizations for nonnormal matrices', Math Comp., 28(127), 679- 693 (1974).Google ScholarCross Ref
- 11.B. N. Parlett, The Symmetric Eigenvalue Problem, Prentice Hall, Englewood Cliffs, New Jersey, 1980. Google ScholarDigital Library
- 12.H. Gazit, 'Randomized parallel connectivity', in Synthesis of Parallel Applications, ed. by J. H. Reif, 197-214, Morgan Kaufmann Publishers, Inc., 1993.Google Scholar
- 13.H. Simon, 'Partitioning unstructured problems for parallel processing', Comput. Syst. Eng., 2(2/3), 135-148 (1991).Google ScholarCross Ref
- 14.Numerical Recipes in C, Second Edition, W. H. Press, S. A. Teukolosky, W. T. Vetterling, and B. P. Flannery, Cambridge University Press, 1994.Google Scholar
Index Terms
- PMRSB: parallel multilevel recursive spectral bisection
Recommendations
OpenMP for Networks of SMPs
In this paper, we present the first system that implements OpenMP on a network of shared-memory multiprocessors. This system enables the programmer to rely on a single, standard, shared-memory API for parallelization within a multiprocessor and between ...
SPMD OpenMP versus MPI on a IBM SMP for 3 Kernels of the NAS Benchmarks
ISHPC '02: Proceedings of the 4th International Symposium on High Performance ComputingShared Memory Multiprocessors are becoming more popular since they are used to deploy large parallel computers. The current trend is to enlarge the number of processors inside such multiprocessor nodes. However a lot of existing applications are using ...
SPMD OpenMP versus MPI on a IBM SMP for 3 Kernels of the NAS Benchmarks
ISHPC '02: Proceedings of the 4th International Symposium on High Performance ComputingShared Memory Multiprocessors are becoming more popular since they are used to deploy large parallel computers. The current trend is to enlarge the number of processors inside such multiprocessor nodes. However a lot of existing applications are using ...
Comments