Provisioning-Based Resource Management for Effective Workflow Scheduling on Utility Grids

摘要

An effective workflow application scheduling on shared resources largely contributes to achieving a high performance in Utility Grids. Users share resources and these resources are autonomously managed in these environments. There is no explicit control on allocating resources to application-tasks on the part of users, the fact that results make users fail in optimizing the application make span and allocation cost. In the current paper, a Minimum First-fit Cost-Make span Trade-off (MinFCMT) heuristic algorithm is developed in order to effectively schedule an application in Utility Grids so that the application make span and allocation-cost can be minimized. To evaluate the MinFCMT heuristic algorithm, widespread simulation of the synthetic workflow is exploited. The results show that the MinFCMT algorithm is more effective than the present algorithms due to optimizing the application make span and allocation-cost in a very low runtime.