Scheduling in a Grid Computing Environment Using Genetic Algorithms

  • Authors:
  • Vincenzo Di Martino;Marco Mililotti

  • Affiliations:
  • -;-

  • Venue:
  • IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

We investigate the possibility to use the computing GRID in a flexible way to permit the maximum usage of resources. In our simulation the jobs submitted by the users provide a characterization of themself to help the system scheduler to do an optimal scheduling in case of resources contention. We use a genetic algorithms to select an optimal or suboptimal scheduling of the jobs. In this preliminary tests we show how the solution founded may maximize the total machine throughput considering not only the single job request but all the job requests during the scheduling process. In this work we show that it is possible to resolve conflicts in the usage of the total computing power and in the data locality for a reasonable number of jobs.