A New Method for Resource Discovery in a Grid Resource Service Chain

Abstract

A new decentralized method is proposed to resource discovery in Grid computing systems. This agent-based method supports semantic description and is flexible for resource identification and allocation. It allows each of the individual resource agents to interact only with its neighbor agents semantically and form a resource chain for a specific task dynamically. We discuss about the success of probability of our method under different task loading rates and compare it with some other known Algorithms. The experiments show the algorithm could flexibly and dynamically discover resources.