Using self-announcer approach for resource availability detection in grid environment

Abstract

Since the Grid is a dynamic environment, the detection of available resources and prediction of resource availability in near future is important for resource scheduling. In a Grid environment, prediction and evaluation of resource availability is the prerequisite for a reasonable resource selection and good scheduling guarantee. There are many approaches for discovery and prediction of the resource availability that have some weaknesses (e.g., complexity time, predicting, using out of date data, etc). In this paper, we present a new method for detection of available resource based on online-announcer with no inquiry from Grid scheduler. We use a rough set analysis in each grid node to get some useful rules for predicting about node's behavior in the grid. The experiment results show that our proposed approach is fast algorithm and it applies effectively a reliable method to predict of proper resources for scheduling.