Resource Discovery and Load Monitoring in Heterogeneous Grid Environment

Abstract

The decentralized broker selects computational re- sources based on actual job requirements, job charac- teristics, and information provided by the resources.A client is a user who submit a job to the server node. A job refers the task which is to be carried out on computational grid. Then job submitted on sched- uler and scheduler will divide a job into smaller tasks and sends that tasks to the available resources.The grid deals with large scale and ever-expanding en- vironment which contains million of users and re- sources. For this reason, resource selection has been a challenging task. This paper presents a method that achieves efficient Grid resource discovery by form- ing and maintaining autonomous resource groups and then monitoring load by using Ganglia monitoring system.