Resource Selection in Computational Grid Based on User QoS and Trust

Grid computing provides a virtual framework for controlled sharing of resources across institutional boundaries. As the resources from different administrative domains participate in task execution, selection of appropriate resources is a challenging task. We propose an efficient method of resource selection for distributed grid environments that is based on execution trust of a resource and the Quality of Service defined for the user. The resources selected are from the trusted list of resources that satisfy the user requirements of computation power for job execution with shorter response time and budget constraints imposed by the user. The selection strategy is based on user QOS parameters which improve the performance of jobs submitted to the grid and the utilization of resources that participate in the Grid. Simulation results show that the proposed model selects the appropriate resources among the available trustworthy resources with an improved power - cost ratio and maximizes the reliability of the resource.