Efficiency of parallel genetic algorithm for solving N-queens problem on multicomputer platform

  • Authors:
  • Milena Lazarova

  • Affiliations:
  • Department of Computer Systems, Technical University of Sofia, Sofia, Bulgaria

  • Venue:
  • EC'08 Proceedings of the 9th WSEAS International Conference on Evolutionary Computing
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

The paper investigates the efficiency of parallel genetic algorithm for solving N-queens problem on a multicomputer platform. The proposed parallel computational model of the genetic algorithm is based on a parallel algorithmic paradigm of synchronous iterations. Dynamic migration of randomly selected chromosomes in a bidirectional circular model is utilized. The algorithm is implemented using both flat (pure MPI) and hybrid (MPI+OpenMP) programming models. The target parallel multicomputer platform is a cluster of SMPs. Performance profiling and scalability analyses have been made in respect of both the workload (board size) and the size of the parallel system.