Optimal data distributions for LU decomposition

  • Thomas Rauber
  • Gudula Rünger
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 966)


The paper considers the well-known problem of LU decomposition to study a method to derive data distributions for parallel computers with a distributed memory organization. The importance of the paper lies not so much in the special application but with the principle that the problem of finding an optimal data distribution is formulated as an optimization problem. This is possible by using a parameterized data distribution and a rigorous performance prediction technique that allows us to derive runtime formulas containing the parameters of the data distribution. The parameters are determined in such a way that the total runtime is minimized, thus also minimizing the communication overhead and the load imbalance penalty.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [BT88]
    D.P. Bertsekas and J.N. Tsitsiklis. Parallel and Distributed Computing. Prentice Hall, New York, NY, 1988.Google Scholar
  2. [DHR94]
    A. Dierstein, R. Hayer, and T. Rauber. A Branch-and-Bound Algorithm for Array Distributions. In 2nd Euromicro Workshop on Parallel and Distributed Processing, pages 528–535, 1994.Google Scholar
  3. [FWM94]
    G.C. Fox, R.D. Williams, and P.C. Messias. Parallel Computing Works! Morgan Kaufmann Publishers, 1994.Google Scholar
  4. [JH89]
    S.L. Johnsson and C.T. Ho. Optimum Broadcasting and Personalized Communication in Hypercubes. IEEE Transactions on Computers, 38(9:1249–1268, 1989.Google Scholar
  5. [RR95]
    T. Rauber and G. Rünger. Performance Predictions for Parallel Diagonal-Implicitly Iterated Runge-Kutta Methods. In 9th Workshop on Parallel and Distributed Simulation, Lake Placid, USA, 1995.Google Scholar
  6. [RRW95]
    T. Rauber, G. Rünger, and R. Wilhelm. An Application Specific Parallel Programming Paradigm. In International Conference on High-Performance Computing and Networking, 1995.Google Scholar
  7. [vdV90]
    E.F. van de Velde. Data redistribution and concurrency. Parallel Computing, 16:125–138, 1990.Google Scholar
  8. [vdV94]
    E.F. van de Velde. Concurrent Scientific Computing. Springer, 1994.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Thomas Rauber
    • 1
  • Gudula Rünger
    • 1
  1. 1.Computer Science Dep.Universität des SaarlandesSaarbrückenGermany

Personalised recommendations