Scheduling framework with resource level load balance using agents in grid computing environments

Aim of the paper is to correlate diverse load-balancing techniques by determining strengths and weaknesses in the considered approaches and thereby propose a novel approach to balance workloads in grids. It presents a comparative study of the implementation and efficiency of various parallel concepts that are common to one or more of the considered techniques. The proposed architecture employs a pattern based technique to determine the type of load which is crucial in bring about load balance and also incorporates a trust agent to indicate the efficiency of the resources available.