Trust Based Load Distribution in a Data Grid Environment

The basic idea of a load balancing system is to transfer partial of system workload from busy nodes to some idle nodes. The target resource selection for task migration is a key factor in the load balance strategy. As grid is dynamic in nature, resources may join or leave the grid at any instance of time. It’s inefficient to select a target resource that often leaves the grid and hence the tasks may be rescheduled for several times. When the cost of load transmission is not negligible, the overhead of the load balancing strategy will be unacceptable. In that case, the cost of propagating system load will be extremely high. To improve the efficiency of the load balancing process, one can select trustworthy resources from the list of available resources. In this paper we present a load distribution strategy that considers the trustworthiness of the resource as a benchmark for participation in the load balancing process. The simulation results show that the load distribution among the trustworthy resources results better system throughput compared to load distribution to all available resources in the grid. Keywords-Data Grid, Load Distribution, Communication Overhead, Trust, Resource Reliabilty