A Distributed Quadtree Index for Peer-to-Peer Settings

Egemen Tanin

CS&SE, Melbourne University

Date and time: 11.30am-12.30pm, Friday 20th August, 2004

Venue: 10.11.04

Chair: Jiankun Hu

Abstract:

Peer-to-peer networks are becoming a common form of online data exchange. Currently, users perform exact-match queries and retrieve complete files on these networks. Many future data intensive applications over P2P settings, e.g., P2P auction networks, P2P job-search networks, and P2P multi-player games, will have to answer complex queries over complex data types. This is currently expected from any web-based application that runs with a database system on a client-server setting. In this talk, I will present a P2P quadtree-based index that enables more powerful accesses to data intensive applications. We have implemented this index for a prototype P2P application and our experiments show that the proposed index is scalable, efficient to use, and exhibits good load-balancing properties. Our work can be applied to various multi-dimensional data types both with spatial and non-spatial attributes.

About the speaker:

Dr. Egemen Tanin has recently joined to the Department of Computer Science and Software Engineering, University of Melbourne, as a lecturer. He has completed his PhD at the University of Maryland at College Park. His areas of interest include spatial data management and distributed data management as well as information visualization. For more information, please see his website at http://www.cs.mu.oz.au/~egemen


Seminar Organisation

Seminars are free and open to the general public. No booking is necessary.

If you are interested in giving a presentation in this seminar series, or to make suggestions for speakers, please contact James Harland, the seminar co-ordinator.

James Harland
Last modified: Fri Aug 13 13:12:52 EST 2004