Mobile agents and the SARA digital library

Remote-sensing data about the Earth's environment is being created at an ever-increasing rate and distributed among heterogeneous remote sites. Traditional models of distributed computing are inadequate to support such complex applications, which generally involve a large quantity of data. We explore an approach based on mobile agent techniques for autonomous data processing and information discovery on the Synthetic Aperture Radar Atlas (SARA) digital library which consists of distributed multi-agency archives of multi-spectral remote-sensing imagery of the Earth. Our goal is to enable automatic and dynamic configuration of distributed parallel computing resources and to efficiently support on-demand processing of such a remote-sensing archive. The design, architecture and implementation of a prototype system that applies this approach is reported.