Neural Approach for Resource Selection with PSO for Grid Scheduling

Grid computing is a computing framework to meet growing demands for running different grid enable applications. This paper proposes a neural network for automatically capturing the requirements of the user and uses it for resource selection. This paper also introduces an approach based on Particle Swarm Optimization (PSO) to schedule jobs on computational grids. The representations of a position and particle velocity in a conventional PSO are extended to real vectors. The proposed approach aims to generate dynamically, an optimal schedule to complete tasks within minimum time duration and also to use resources efficiently.