Parallel vertex cover: a case study in dynamic load balancing

  • Authors:
  • Dinesh P. Weerapurage;John D. Eblen;Gary L. Rogers, Jr.;Michael A. Langston

  • Affiliations:
  • University of Tennessee, Knoxville TN;University of Tennessee, Knoxville TN;University of Tennessee, Knoxville TN;University of Tennessee, Knoxville TN

  • Venue:
  • AusPDC '11 Proceedings of the Ninth Australasian Symposium on Parallel and Distributed Computing - Volume 118
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

The significance of dynamic parallel load balancing is considered in the context of fixed parameter tractability. The well known vertex cover problem is used as a case study. Several algorithms are developed and tested on graphs derived from real biological data. Implementations are carried out on the Kraken supercomputer, currently the world's fastest computational platform for open science. We show that for certain difficult instances of biological data graphs our approach scales well up to 2400 processors.