The University of Texas at Austin
Harrick Vin home pageComputer Sciences

Research  »  Completed Projects  »  Trellis


 Project Summary

The growth of the Internet has enabled users to access information stored at geographically distributed sites. The non-uniformity of information access, however, creates hot-spots of network and server load, which significantly increase the latency for information access.

The Trellis project addresses the scalability requirements of emerging distributed applications by developing: (1) scalable servers that can support a large number of connections and I/O requests while guaranteeing data availability, and (2) scalable distributed cache architecture.

 Team Members

Renu Tewari, Jonathan Kay, Michael Dahlin (»)

 Project Publications

Titles link to the abstract. Use pdf icons in the right margin to download the publication

Design Considerations for Distributed Caching on the Internet
R. Tewari, M. Dahlin, H.M. Vin, and J. Kay
Proceedings, International Conference on Distributed Computing and Systems (ICDCS), May 1999, pp. 273-284.

Resource-based Caching for Web Servers
R. Tewari, H.M. Vin, A. Dan, and D. Sitaram
Proceedings, ACM/SPIE Multimedia Computing and Networking 1998 (MMCN'98), San Jose, January 1998, pp. 191-204. (ps)

Design and Performance Tradeoffs in Clustered Video Servers
R. Tewari, R. Mukherjee, D.M. Dias, and H.M. Vin
Proceedings, IEEE International Conference on Multimedia Computing and Systems 1996 (ICMCS'96), Tokyo, Japan, May 1996, pp. 144-150. (ps)

High Availability in Clustered Multimedia Servers
R. Tewari, D.M. Dias, R. Mukherjee, and H.M. Vin
Proceedings, IEEE International Conference on Data Engineering, New Orleans, February 1996, pp. 345-354. (ps)

To top of page To top of page