Improvement of the Efficiency of Genetic Algorithms for Scalable Parallel Graph Partitioning in a Multi-level Framework

Abstract

Parallel graph partitioning is a difficult issue, because the best sequential graph partitioning methods known to date are based on iterative local optimization algorithms that do not parallelize nor scale well. On the other hand, evolutionary algorithms are highly parallel and scalable, but converge very slowly as problem size increases. This paper presents methods that can be used to reduce problem space in a dramatic way when using graph partitioning techniques in a multi-level framework, thus enabling the use of evolutionary algorithms as possible candidates, among others, for the realization of efficient scalable parallel graph partitioning tools. Results obtained on the recursive bipartitioning problem with a multi-threaded genetic algorithm are presented, which show that this approach outperforms existing state-of-the-art parallel partitioners.KeywordsGenetic AlgorithmMemetic AlgorithmLarge GraphGraph PartitioningNest DissectionThese 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.