Parallel hypergraph partitioning for scientific computing

  • Authors:

  • Karen D. Devine;Erik G. Boman;Robert T. Heaphy;Rob H. Bisseling;Umit V. Catalyurek

  • Affiliations:

  • Sandia National Laboratories, Dept. of Discrete Algorithms and Math., Albuquerque, NM;Sandia National Laboratories, Dept. of Discrete Algorithms and Math., Albuquerque, NM;Sandia National Laboratories, Dept. of Discrete Algorithms and Math., Albuquerque, NM;Utrecht University, Dept. of Mathemathics, Utrecht, The Netherlands;Ohio State University, Dept. of Biomedical Informatics, Columbus, OH

  • Venue:
  • IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Graph partitioning is often used for load balancing in parallel computing, but it is known that hypergraph partitioning has several advantages. First, hypergraphs more accurately model communication volume, and second, they are more expressive and can better represent nonsymmetric problems. Hypergraph partitioning is particularly suited to parallel sparse matrix-vector multiplication, a common kernel in scientific computing. We present a parallel software package for hypergraph (and sparse matrix) partitioning developed at Sandia National Labs. The algorithm is a variation on multilevel partitioning. Our parallel implementation is novel in that it uses a two-dimensional data distribution among processors. We present empirical results that show our parallel implementation achieves good speedup on several large problems (up to 33 million nonzeros) with up to 64 processors on a Linux cluster.