Abstract
A number of techniques are described for solving sparse linear systems on parallel platforms. The general approach used is a domai n-decomposition type method in which a processor is assigned a certain number of rows of the linear system to be solved. Strategies that are discussed include non-standard graph partitioners, and a forced loadbalance technique for the local iterations. A common practice when partitioning a graph is to seek to minimize the number of cut-edges and to have an equal number of equations per processor. It is shown that partitioners that take into account the values of the matrix entries may be more effective.
Work supported by NSF under grant CCR-9618827, and in part by the Minnesota Supercomputer Institute.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
X. C. Cai and Y. Saad. Overlapping domain decomposition algorithms for general sparse matrices. Numerical Linear Algebra with Applications, 3:221–237, 1996.
I. S. Duff, R. G. Grimes, and J. G. Lewis. Sparse matrix test problems. ACM Transactions on Mathematical Software, 15:1–14, 1989.
V. Eijkhout and T. Chan. ParPre a parallel preconditioners package, reference manual for version 2.0.17. Technical Report CAM Report 97-24, UCLA, 1997.
C. Farhat and M. Lesoinne. Mesh partitioning algorithms for the parallel solution of partial differential equations. Applied Numerical Mathematics, 12, 1993.
J. A. George and J. W. Liu. Computer Solution of Large Sparse Positive Definite Systems. Prentice-Hall, Englewood Cliffs, NJ, 1981.
T. Goehring and Y. Saad. Heuristic algorithms for automatic graph partitioning. Technical Report UMSI 94-29, University of Minnesota Supercomputer Institute, Minneapolis, MN, February 1994.
B. Hendrickson and R. Leland. An improved spectral graph partitioning algorithm for mapping parallel computations. Technical Report SAND92-1460, UC-405, Sandia National Laboratories, Albuquerque, NM, 1992.
Scott A. Hutchinson, John N. Shadid, and R. S. Tuminaro. Aztec user’s guide. version 1.0. Technical Report SAND95-1559, Sandia National Laboratories, Albuquerque, NM, 1995.
M. T. Jones and P. E. Plassmann. BlockSolve95 users manual: Scalable library software for the solution of sparse linear systems. Technical Report ANL-95/48, Argonne National Lab., Argonne,IL., 1995.
G. Karypis. Graph Partitioning and its Applications to Scientific Computing. PhD thesis, Department of Computer Science, University of Minnesota, Minneapolis, MN, 1996.
S. Kuznetsov, G. C. Lo, and Y. Saad. Parallel solution of general sparse linear systems. Technical Report UMSI 97/98, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, MN, 1997.
J. W. H. Liu. A graph partitioning algorithm by node separators. ACM Transactions on Mathematical Software, 15:198–219, 1989.
A. Pothen, H. D. Simon, and K. P. Liou. Partitioning sparse matrices with eigenvectors of graphs. SIAM Journal on Matrix Analysis and Applications, 11:430–452, 1990.
Y. Saad. Parallel sparse matrix library (P_SPARSLIB): The iterative solvers module. In Advances in Numerical Methods for Large Sparse Sets of Linear Equations, Number 10, Matrix Analysis and Parallel Computing, PCG 94, pages 263–276, Keio University, Yokohama, Japan, 1994.
Y. Saad. Iterative Methods for Sparse Linear Systems. PWS publishing, New York, 1996.
Y. Saad and A. Malevsky. PSPARSLIB: A portable library of distributed memory sparse iterative solvers. In V. E. Malyshkin et al., editor, Proceedings of Parallel Computing Technologies (PaCT-95), 3-rd international conference, St. Petersburg, Russia, Sept. 1995, 1995.
Y. Saad and M. Sosonkina. Distributed Schur complement techniques for general sparse linear systems. Technical Report UMSI 97/159, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, MN, 1997. Submitted, Revised.
B. Smith, P. Bjørstad, and W. Gropp. Domain decomposition: Parallel multilevel methods for elliptic partial differential equations. Cambridge University Press, New-York,NY, 1996.
B. Smith, W. D. Gropp, and L. C. McInnes. PETSc 2.0 user’s manual. Technical Report ANL-95/11, Argonne National Laboratory, Argonne, IL, July 1995.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Saad, Y., Sosonkina, M. (1999). Non-standard Parallel Solution Strategies for Distributed Sparse Linear Systems. In: Zinterhof, P., Vajteršic, M., Uhl, A. (eds) Parallel Computation. ACPC 1999. Lecture Notes in Computer Science, vol 1557. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-49164-3_2
Download citation
DOI: https://doi.org/10.1007/3-540-49164-3_2
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65641-8
Online ISBN: 978-3-540-49164-4
eBook Packages: Springer Book Archive