Grid Environment, Resource Discovery Mechanism

Grid computing is a fast developing wide area network computing technique. Its objective is to combine Internet into a whole very large-scale computer system in order to share all computing resources, storage resources, information resources and knowledge resources. Resource discovery is an important research problem, which mainly solves the problem that how to discover the relevant resources according to clients' requests for resources.In order to provide clients with a scalable, flexible environment for resource discovery, this paper puts forward a system model for resource discovery based on administrative domains according to some features of resources in grid environment, such as geographical dispersion ,dynamic change, and heterogeneous systems. Furthermore, this paper also introduces the composing , work process in this system and some critical problems which need to be solved in it, meanwhile, several key issues such as method for expressing information, method for dynamic discovery and matching mechanism in the same administrative domain are analyzed and discussed in focus.Resource location is the kernel problem that should be studied in grid resource discovery system. It's mostly response for the routing transfer of resources request and the matching of resources information, the performance of which has great impact on the efficiency of the whole resource discovery system. In grid environment, given that not only locality information of resource, but also the semantic information of multi-types and multi-attributes of resources should be considered in resource location. This paper puts forward the ideas of range queries of numerical attributes and multidimensional queries. Furthermore, with these query ideas, a scalable resource location method single attribute dominated query method is provided on the basis of the location method of Chord. And Simulative experimental results also indicate that this method has good scalability.