Application Research Of Grid Information Interoperability Based On Agent And Ontology

As the third information technology period after Internet and Web, Grid technology uses web services as reference. However, it is difficult to spread grid technology since grid is highly heterogeneous, cross-platform and dynamic. There are many problems remained in grid and one of the most important is the interoperability in grid information.To solve the interoperability problem, we build a grid information interoperability model based on agents and ontology to integrate their characteristics and advantages. The model discusses how to make grid information services more interoperable in the way that we utilize the advantage of agents and ontology to overcome the existing deficiencies of web services in Grid computing and application pattern. On one hand, we can process dynamic and distributed grid resources by combining mobile, autonomy agents. On the other hand, we can solve the semantic problem in grid information by using mature ontology technology to describe resources. It will give us a good sample in solving interoperability problem in grid information after integrating agents and ontology into Grid technology.Our work mainly focuses on the following aspects:(1) Introduce the idea of Agent and Ontology to slove the interoperability problem.(2) Describe agent technology systematically, discuss the application of agents in grid information, especially the management of distributed grid resources and agent-based Grid inter-connection structure model. We also describe ontology and its application, and then illustrate how to use OWL-S to describe Grid services and many semantic convert methods in Grid services.(3) Utilize the advantage of agents and ontology to build grid information interoperability model based on agents and ontology, analysis the structure and advantage of the model, finally introduce some related technologies.(4) Finally, we design a prototype system—grid information query system, which includes local libraries to describe grid resources, personalized agents to process information and many system interfaces.