Partitioning sparse rectangular matrices for parallel processing

  • Tamara G. Kolda
Regular Talks
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1457)

Abstract

We are interested in partitioning sparse rectangular matrices for parallel processing. The partitioning problem has been well-studied in the square symmetric case, but the rectangular problem has received very little attention. We will formalize the rectangular matrix partitioning problem and discuss several methods for solving it. We will extend the spectral partitioning method for symmetric matrices to the rectangular case and compare this method to three new methods — the alternating partitioning method and two hybrid methods. The hybrid methods will be shown to be best.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Charles J. Alpert and So-Zen Yao. Spectral partitioning: The more eigenvectors, the better. In 32nd ACM/IEEE Design Automation Conference, pages 195–200, 1995.Google Scholar
  2. [2]
    Stephen T. Barnard and Horst D. Simon. A fast multilevel implementation of recursive spectral bisection for partitioning unstructured problems. Concurrency: Practice and Experience, 6:101–117, 1994.Google Scholar
  3. [3]
    Michael W. Berry, Bruce Hendrickson, and Padma Raghavan. Sparse matrix reordering schemes for browsing hypertext. In James Renegar, Michael Shub, and Steve Smale, editors, The Mathematics of Numerical Analysis, volume 32 of Lectures in Applied Mathematics, pages 99–122. American Mathematical Society, 1996.Google Scholar
  4. [4]
    Julie Falkner, Franz Rendl, and Henry Wolkowicz. A computational study of graph partitioning. Math. Prog., 66:211–239, 1994.Google Scholar
  5. [5]
    Miroslav Fiedler. Algebraic connectivity of graphs. Czechoslovak Mathematical J., 23:298–305, 1973.Google Scholar
  6. [6]
    Michael R. Garey and David S. Johnson. Computers and Intractability: A Guide to the Theory of NP-Completeness. W. H. Freeman and Company, New York, 1979.Google Scholar
  7. [7]
    Alan George and Joseph W. Liu. Computer Solution of Large Sparse Positive Definite Systems. Prentice-Hall Series in Computational Mathematics. Prentice-Hall, Englewood Cliffs, 1981.Google Scholar
  8. [8]
    Stephen Guattery and Gary L. Miller. On the quality of spectral seperators. Accepted for publication in SIAM J. Matrix Anal. Appl., 1997.Google Scholar
  9. [9]
    Bruce Hendrickson and Tamara G. Kolda. Partitioning nonsquare and nonsymmetric matrices for parallel processing. In preparation, 1998.Google Scholar
  10. [10]
    Bruce Hendrickson and Tamara G. Kolda. Partitioning sparse rectangular matrices for parallel computations of Ax and A T v. Accepted for publication in Proc. PARA98: Workshop on Applied Parallel Computing in Large Scale Scientific and Industrial Problems, 1998.Google Scholar
  11. [11]
    Bruce Hendrickson and Robert Leland. Multidimensional spectral load balancing. Technical Report 93-0074, Sandia Natl. Lab., Albuquerque, NM, 87185, 1993.Google Scholar
  12. [12]
    Bruce Hendrickson and Robert Leland. An improved spectral graph partitioning algorithm for mapping parallel computations. SIAM J. Sci. Stat. Comput., 16:452–469, 1995.Google Scholar
  13. [13]
    Bruce Hendrickson and Robert Leland. A multilevel algorithm for partitioning graphs. In Proc. Supercomputing '95. ACM, 1995.Google Scholar
  14. [14]
    Bruce Hendrickson, Robert Leland, and Rafael Van Driessche. Skewed graph partitioning. In Proc. Eighth SIAM Conf. on Parallel Processing for Scientific Computing. SIAM, 1997.Google Scholar
  15. [15]
    George Karypis and Vipin Kumar. A fast and high quality multilevel scheme for paritioning irregular graphs. Technical Report 95-035, Dept. Computer Science, Univ. Minnesota, Minneapolis, MN 55455, 1995.Google Scholar
  16. [16]
    George Karypis and Vipin Kumar. Parallel multilevel graph partitioning. Technical Report 95-036, Dept. Computer Science, Univ. Minnesota, Minneapolis, MN 55455, 1995.Google Scholar
  17. [17]
    Tamara G. Kolda. Limited-Memory Matrix Methods with Applications. PhD thesis, Applied Mathematics Program, Univ. Maryland, College Park, MD 20742, 1997.Google Scholar
  18. [18]
    Tamara G. Kolda and Dianne P. O'Leary. A semi-discrete matrix decomposition for latent semantic indexing in information retrieval. Accepted for publication in ACM Trans. Information Systems, 1997.Google Scholar
  19. [19]
    Dianne P. O'Leary and Shmuel Peleg. Digital image compression by outer product expansion. IEEE Trans. Comm., 31:441–444, 1983.Google Scholar
  20. [20]
    Christopher C. Paige and Michael A. Saunders. LSQR: An algorithm for sparse linear equations and sparse least squares. ACM Trans. Mathematical Software, 8:43–71, 1982.Google Scholar
  21. [21]
    Alex Pothen, Horst D. Simon, and Kang-Pu Liou. Partitioning sparse matrices with eigenvectors of graphs. SIAM J. Matrix Anal. Appl., 11:430–452, 1990.Google Scholar
  22. [22]
    Horst D. Simon. Partitioning of unstructured problems for parallel processing. In Computing Systems in Engineering, number 2/3, pages 135–148. Pergammon Press, 1991.Google Scholar
  23. [23]
    Daniel A. Speilman and Shang-Hua Teng. Spectral partitioning works: Planar graphs and finite element meshes. Unpublished manuscript, 1996.Google Scholar

Copyright information

© Springer-Verlag 1998

Authors and Affiliations

  • Tamara G. Kolda
    • 1
  1. 1.Computer Science and Mathematics DivisionOak Ridge National LaboratoryOak Ridge

Personalised recommendations