Advertisement

A Parallelisable Multi-level Banded Diffusion Scheme for Computing Balanced Partitions with Smooth Boundaries

  • François Pellegrini
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4641)

Abstract

Graph partitioning algorithms have yet to be improved, because graph-based local optimization algorithms do not compute smooth and globally-optimal frontiers, while global optimization algorithms are too expensive to be of practical use on large graphs. This paper presents a way to integrate a global optimization, diffusion algorithm in a banded multi-level framework, which dramatically reduces problem size while yielding balanced partitions with smooth boundaries. Since all of these algorithms do parallelize well, high-quality parallel graph partitioners built using these algorithms will have the same quality as state-of-the-art sequential partitioners.

Keywords

Global Optimization Algorithm Graph Vertex Test Graph Full Graph Balance Partition 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Hendrickson, B., Leland, R.: A multilevel algorithm for partitioning graphs. In: Proceedings of Supercomputing (1995)Google Scholar
  2. 2.
    Karypis, G., Kumar, V.: A fast and high quality multilevel scheme for partitioning irregular graphs. SIAM J. on Scientific Computing 20(1), 359–392 (1998)CrossRefMathSciNetGoogle Scholar
  3. 3.
    Kernighan, B.W., Lin, S.: An efficient heuristic procedure for partitionning graphs. BELL System Technical Journal 49, 291–307 (1970)Google Scholar
  4. 4.
    Fiduccia, C.M., Mattheyses, R.M.: A linear-time heuristic for improving network partitions. In: Proc. 19th Design Automat. Conf. pp. 175–181. IEEE Computer Society Press, Los Alamitos (1982)Google Scholar
  5. 5.
    Vanderstraeten, R., Keunings, R., Farhat, C.: Beyond conventional mesh partitioning algorithms. In: SIAM Conf. on Par. Proc. pp. 611–614 (1995)Google Scholar
  6. 6.
    Diekmann, R., Preis, R., Schlimbach, F., Walshaw, C.: Aspect ratio for mesh partitioning. In: Pritchard, D., Reeve, J.S. (eds.) Euro-Par 1998. LNCS, vol. 1470, pp. 347–351. Springer, Heidelberg (1998)CrossRefGoogle Scholar
  7. 7.
    Meyerhenke, H., Schamberger, S.: Balancing parallel adaptive FEM computations by solving systems of linear equations. In: Cunha, J.C., Medeiros, P.D. (eds.) Euro-Par 2005. LNCS, vol. 3648, pp. 209–219. Springer, Heidelberg (2005)Google Scholar
  8. 8.
    Meyerhenke, H., Schamberger, S.: A parallel shape optimizing load balancer. In: Nagel, W.E., Walter, W.V., Lehner, W. (eds.) Euro-Par 2006. LNCS, vol. 4128, pp. 232–242. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  9. 9.
    Wan, Y., Roy, S., Saberi, A., Lesieutre, B.: A stochastic automaton-based algorithm for flexible and distributed network partitioning. In: Proc. Swarm Intelligence Symposium, pp. 273–280. IEEE, Los Alamitos (2005)CrossRefGoogle Scholar
  10. 10.
    Chevalier, C., Pellegrini, F.: Improvement of the efficiency of genetic algorithms for scalable parallel graph partitioning in a multi-level framework. In: Nagel, W.E., Walter, W.V., Lehner, W. (eds.) Euro-Par 2006. LNCS, vol. 4128, pp. 243–252. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  11. 11.
    Chevalier, C., Pellegrini, F.: PT- SCOTCH: A tool for efficient parallel graph ordering. Parallel Computing (submitted), http://www.labri.fr/~pelegrin/papers/scotch_parallelordering_parcomp.pdf
  12. 12.
    SCOTCH: Static mapping, graph partitioning, and sparse matrix block ordering package, http://www.labri.fr/~pelegrin/scotch/
  13. 13.
    Simon, H.D., Teng, S.H.: How good is recursive bipartition. SIAM J. Scientific Computing 18(5), 1436–1445 (1997)zbMATHCrossRefMathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • François Pellegrini
    • 1
  1. 1.ENSEIRB, LaBRI and INRIA Futurs, Université Bordeaux I, 351, cours de la Libération, 33405 TALENCEFrance

Personalised recommendations