ABSTRACT

Static task scheduling in computational grids is very important because of the optimal usage of computing time for scheduling algorithms. Given a set of resources, a static scheduler computes the execution schedule before runtime. In static task scheduling resource information and performance parameters are assumed to be known depending on how a job can be divided, relevant research can be categorized into two different areas: divisible workload, scheduling where they can divided workload into arbitraty-sizes. Solving this problem dynamically needs more time. Therefore an attempt is made to solve it by meta-heuristic techniques. A new GA scheduler, GASAScheduler is presented whose run-time depends on the number of tasks in scheduling problem. The computation time to find sub-optimal function improved. The result shows the computation time of the proposed algorithms is better.