Orchestrating Provisioning among Heterogeneous Grid Middleware

Abstract

After several years development, the computational infrastructure has been widely developed. Especially, when the grid technology grows to production level, users have several options to handle resource less problems when submitting large scale jobs to the infrastructure. Capabilities of dynamic provisioning, fast deployment of virtual machines, and interoperation can help users finishing their jobs correctly even the local resources are not adequate. To orchestrate these approaches well and find the optimal solution for users, this paper proposes a novel orchestrating model to optimize these approaches for high efficiency. An image processing application running on three heterogeneous grid platforms is adopted to demonstrate the efficiency and capability of proposed model. The result proved that the optimal solution can efficiently orchestrate the jobs among the grids.