Research of a massive distributed remote sensing data resource sharing method under grid environment

With the rapid growth of the volume of remote sensing data, there is no doubt that resource sharing has become the key factor to improve the utilization of remote sensing data. This paper discusses and compares four major approaches of remote sensing data resource sharing, studies the theories of Grid and Grid middleware, and proposes an effective method of remote sensing data resource sharing under Grid environment. Firstly, by analyzing the features of Grid environment and remote sensing data, the metadata models are constructed and expressed using XML technology. Secondly, a resource directory service is structured for registering, browsing, inquiring and managing remote sensing metadata by adopting RLS, which is a data resource management component of Globus Toolkit. Based on the above study, in order to transmit data transparently, reliably and safely, this paper explores a reliable transmission method by RFT and GridFTP. Finally, a simulation Grid environment is built and a Grid middleware is implemented using object-oriented design method and reusable software design idea. Our proposed approach fully takes advantage of the Grid characteristic of maximizing the resource sharing. Furthermore, the middleware discussed here effectively improves the security of resource sharing, heterogeneous resources integration and has an easy operation. And most of all, this study provides a valid method and technical support for improving the utilization efficiency of remote sensing data.