Research On Grid-Based Information Integration System

With the development of grid technology, information resources varied from traditional structured resources to semi-structured resources, even unstructured resources. It's difficult for users to access and query information uniformly and transparently in grid, so information integration in grid is needed.Information integration provides a uniform and transparent access, representation, query and management for resources. This makes it an effective way to integrate heterogeneous resources in grid environment. Information integration can be implemented with two approaches: 1) materialized approach; 2) virtual approach. Because resources in Grid are heterogeneous and dynamic, virtual approach is a sharp tool of integrating resources in grid Environment; but the network communication is a big constraint on system performance improvement.In this thesis, Multi-Agent technology is applied into our information integration system to improve system performance, and a performance optimize strategy based on resource clustering is proposed. Based on the performance optimize strategy, a resource replication strategy is proposed to balance the network load of resource transmition and improve the system query performance.