A Policy-Based Adaptive Scheduling Model of Software Resource in ITSGrid

Abstract

Because of the distribution and heterogeneity of application software license resource in ITSGrid, more than one resource node could satisfy requirements of large-scale computing services in ITSGrid. How to select software license resources among ITS resource nodes, determine appropriate license resource scheduling strategy for computing tasks and maximize utilization rate of expensive large-scale software is becoming a problem must be addressed in ITSGrid. In this paper, we analyze the architecture of resource node in ITSGrid, design a policy-based adaptive scheduling model of application software license resource for ITSGrid, propose relative adaptive scheduling algorithm for application software license, solve license resource global scheduling problem in ITSGrid, which could improve ITSGrid system throughput, shorten waiting and response time of large-scale computing jobs and maximize benefit of expensive application software in ITSGrid.