Supporting next-generation distributed applications

Abstract

Even the current euphoria over the World Wide Web doesn't do full justice to the Internet's potential. With the manifold increase in CPU processing power and network bandwidth, it's inevitable that the Internet will support increasingly complex distributed applications. While information retrieval applications dominate the Internet today, the next-generation Internet will likely offer applications that can process massive amounts of data for visualization and support real-time interactivity. For instance, a digital library of satellite imagery might be processed for feature extraction or visualization. A virtual environment for training fire fighters might involve distributed simulations and real-time user interactivity. In 1994, the University of Texas at Austin opened the Distributed Multimedia Computing Laboratory (DMCL) to conduct the basic and experimental research necessary to address the problems of these emerging applications. The research work done at DMCL can be broadly classified into the following categories: resource management techniques that meet the performance requirements of applications (the Symphony, NetCop, and OSng projects); an information delivery architecture that meets the scalability requirements of applications (the Trellis project); a fault-tolerance framework for cooperative distributed applications (the Coyote project); an environment for creating and disseminating digital educational material over the Internet (the InfoWeave project). The article summarizes key research findings and ongoing projects at DMCL.