An Efficient Semantic Discovery of Grid Resources

Grid computing provides huge number of hardware and software resources. For efficient discovery of Grid resources, it is necessary to consider resource usage policies during matchmaking process, since the discovered resources are useless if they are unavailable to use. In this paper we propose a framework for efficient discovery of Grid resources. It contains a semantic knowledge base, in the form of domain, functional and policy Ontologies, of resources description. Functional ontology assists to control operations that user want to perform with resources. We use policy ontology to enforce user specified, resource usage policies. Moreover, we also develop a policy analyzer algorithm to enforce these policies in matchmaking process. Our objective is to make prior resource selection efforts more useful by providing maximum surety that the discovered resources are available to the user by analyzing their usage policies. A request subscription knowledge base is also maintained which is used to check the request and send notification later on, if resources are not available. This framework will be evaluated using ProActive Grid Middleware.