Parallel graph-partitioning using the mob heuristic
In this paper we study graph partitioning which is an important NP-complete problem. We have implemented two mob heuristics and analysed their performance. Our main result is that for randomly generated graphs the mob heuristics with an exponential schedule is more efficient than that with the linear schedule, proposed earlier. With the other parameters of the heuristic being identical the exponential schedule doubles the speedup.
KeywordsPVM graph partitioning heuristic SP-2
Unable to display preview. Download preview PDF.
- 1.M. R. Garey and D. S. Johnson. Computers and Intractability: A Guide to the Theory of NP-Completeness. W. H. Freeman, San Francisco, 1979.Google Scholar
- 3.B. W. Kernighan and S. Lin. An effecient heuristic procedure for partitioning graphs. AT & T Bell Labs. Tech. J., 49:291–307, 1970.Google Scholar
- 4.S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi. Optimization by simulated annealing. Science, 220:671–680, 1983.Google Scholar
- 6.M. Šoch, J. Trdlička, and P. Tvrdć. PVM, computational geometry, and parallel computing course. In A. Bode, J. Dongarra, T. Ludwig, and V. Sunderam, editors, Parallel Virtual Machine — EuroPVM'96, number 1156 in Lecture Notes of Computer Science, pages 38–44. Springer Verlag, 1996.Google Scholar