Abstract

When a large number of tasks request different resources in a grid system, resources allocation should have been done with proper planning and scheduling to guarantee a good quality of service (QOS). There are different ways to provide these requests by choosing appropriate resources allocation to optimize the total operation of the system. In this study, first some parameters, such as priority, delay, reliability and cost are determined for each task to maximize system performance and appropriate resources distribution. Then, a hybrid optimization algorithm for choosing grid resource based on genetic algorithm and particle swarm optimization (PSO) is presented. Based on the experimental results, this method’s performance is 17.5% more than genetic and PSO algorithms in average.

Key words: Grid system, resources selection, quality of service, genetic algorithm, particle swarm optimization (PSO).