A Three-Stage Approach for grid task scheduling

Grid computing has the capability for solving complex scientific applications. The complex applications are divided into smaller units to make the process easier. Smaller units are scheduled in such a manner that the overall cost is minimized. So, scheduling is an important part of Grid Environment. The aim of scheduling is proper utilization of resources and minimizes the completion time. In task scheduling, the job is divided into number of Meta tasks. Meta task do not have any dependency between each other. Now, the problem is to map Meta tasks onto resources such that the Makespan is minimized. In this paper, we propose a Three Stage Approach (TSA) for grid task scheduling. The stages are threshold, queuing using priority assignment and resource selection. Our experimental results and analysis shows better results than traditional scheduling algorithms like Minimum Execution Time (MET), Minimum Completion Time (MCT), Min-Min and Max-Min. Our approach provides optimum solution or a solution close to optimum.