Skip to Main Content
To achieve the best quality of overall optimal performance on grid environment and user services, job loads need to be balanced in grids due to the non-uniformly distributed nature of jobs submitted to grids. Jobs are transferred from a heavy loaded grid to other grids in order to achieve performance optimization. The running cost of a job is calculated at the time it is completed. In this paper, we propose a job schedule model considering job cost and performance. The algorithm is resided in all grids to schedule and arrange their own jobs by themselves to get best performance without extra user cost.