A Framework for Integrating Network Information into Distributed Iterative Solution of Sparse Linear Systems

  • Devdatta Kulkarni
  • Masha Sosonkina
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2565)

Abstract

Recently, we have proposed a design of an easy-to-use network information discovery tool that can interface with a distributed application non-intrusively and without incurring much overhead. The application is notified of the network changes in a timely manner and may react to the changes by invoking the adaptation mechanisms encapsulated in notification handlers. Here we describe possible adaptations of a commonly used scientific computing kernel, distributed sparse largescale linear system solution code.

References

  1. [1]
    D. Andersen, D. Bansal, D. Curtis, S. Seshan, and H. Balakrishnan. System support for bandwidth management and content adaptation in Internet applications. In Proceedings of 4th Symposium on Operating Systems Design and Implementation San Diego, CA, October 2000. USENIX Association.:213–226, 2000. 437Google Scholar
  2. [2]
    J. Dongarra, G. Fagg, A. Geist, and J. A. Kohl. HARNESS: Heterogeneous adaptable reconfigurable NEtworked systems. pages 358–359, http://citeseer.nj.nec.com/327665.html, 1998. 437
  3. [3]
    X. Fu, W. Shi, A. Akkerman, and V. Karamcheti. CANS: Composable, adaptive network services infrastructure. In Proceedings of the USENIX Symposium on Internet Technologies and Systems, 2001. 437Google Scholar
  4. [4]
    D. Gunter, B. Tierney, B. Crowley, M. Holding, and J. Lee. Netlogger: A toolkit for distributed system performance analysis. In Proceedings of the IEEE Mascots 2000 Conference, 2000. 437Google Scholar
  5. [5]
    D. Kulkarni and M. Sosonkina. Using dynamic network information to improve the runtime performance of a distributed sparse linear system solution. Technical Report UMSI-2002-10, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, MN, 2002. accepted in VECPAR 2002. 439Google Scholar
  6. [6]
    Z. Li, Y. Saad, and M. Sosonkina. pARMS: A parallel version of the algebraic recursive multilevel solver. Technical Report UMSI-2001-100, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, MN, 2001. 437, 439, 440, 445Google Scholar
  7. [7]
    B. Lowekamp, N. Miller, D. Sutherland, T. Gross, P. Steenkiste, and J. Subhlok. A resource query interface for network-aware applications. Cluster Computing, 2:139–151, 1999. 437CrossRefGoogle Scholar
  8. [8]
    R. T. Mills, A. Stathopoulos, and E. Smirni. Algorithmic modifications to the Jacobi-Davidson parallel eigensolver to dynamically balance external CPU and memory load. In Proceedings of the International Conference on Supercomputing 2001,Sorr ento, Italy, pages 454–463, June 18–22, 2001. 437Google Scholar
  9. [10]
    Y. Saad. Iterative Methods for Sparse Linear Systems. PWS publishing, New York, 1996. 439, 440, 443, 444MATHGoogle Scholar
  10. [11]
    Y. Saad and M. Sosonkina. Distributed Schur Complement techniques for general sparse linear systems. SIAM J. Scientific Computing, 21(4):1337–1356, 1999. 439CrossRefMathSciNetGoogle Scholar
  11. [12]
    Y. Saad and M. Sosonkina. Non-standard parallel solution strategies for distributed sparse linear systems. In A. Uhl P. Zinterhof, M. Vajtersic, editor, Parallel Computation: Proc. of ACPC’99, Lecture Notes in Computer Science, Berlin, 1999. Springer-Verlag. 441, 443Google Scholar
  12. [13]
    B. Smith, P. Bjørstad, and W. Gropp. Domain Decomposition: Parallel Multilevel Methods for Elliptic Partial Differential Equations. Cambridge University Press, New York, 1996. 439, 441MATHGoogle Scholar
  13. [14]
    M. Snir, S. Otto, S. Huss-Lederman, D. Walker, and J. Dongarra. MPI-The complete Reference, volume 1. The MIT Press, second edition, 1998. 436Google Scholar
  14. [15]
    M. Sosonkina. Runtime adaptation of an iterative linear system solution to distributed environments. In Applied Parallel Computing,P ARA’2000, volume 1947 of Lecture Notes in Computer Science, pages 132–140, Berlin, 2001. Springer-Verlag. 441, 443, 444Google Scholar
  15. [16]
    M. Sosonkina and G. Chen. Design of a tool for providing network information to distributed applications. In Parallel Computing Technologies PACT2001, volume 2127 of Lecture Notes in Computer Science, pages 350–358. Springer-Verlag, 2001. 437CrossRefGoogle Scholar
  16. [17]
    C. Wagner. Introduction to algebraic multigrid-course notes of an algebraic multigrid. University of Heidelberg 1998/99. 440Google Scholar
  17. [18]
    R. Wolski. Dynamically forecasting network performance using the network weather service. Cluster Computing, 1(1):119–132, 1998. 437CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Devdatta Kulkarni
    • 1
  • Masha Sosonkina
    • 2
  1. 1.Department of Computer ScienceUniversity of MinnesotaMinneapolisUSA
  2. 2.Department of Computer ScienceUniversity of MinnesotaDuluthMNUSA

Personalised recommendations