MULTIOBJECTIVE EVOLUTIONARY ALGORITHMS FOR SCHEDULING JOBS ON COMPUTATIONAL GRIDS
Crina Groşan, Ajith Abraham, Bjarne E. Helvik · 2007
In a computational grid, at time t, the task is to allocate the user defined jobs efficiently by meeting the deadlines and making use of all the available resources. In the past, objectives were combined and the problem is very often simplified to a single objective problem. In this paper, we formulate a novel Evolutionary Multi-Objective (EMO) approach by using the Pareto dominance and the objectives are formulated independently. We report some preliminary experiments and the performance of the EMO approach is compared with simulated annealing and particle swarm optimization techniques. Empirical results indicate that the proposed EMO approach is very efficient.