Statistical Clustering of Internet Communication Patterns

F. Hernendez-Campos, A. Nobel, K. Jeffay, F.D. Smith
The 35th Symposium on the Interface of Computing Science and Statistics, Salt Lake City, UT, July 2003, to appear.

ABSTRACT: We describe a new methodology for analyzing Internet traffic based on an abstract model of application-level communication and statistical cluster analysis, and argue that this method of analysis can serve as a foundation for building flexible traffic generation tools. We present the details of two case studies in which the new analysis tools are applied to data from the University of North Carolina at Chapel Hill and Internet2.

Get a PostScript (compressed) or PDF copy of the paper.

Back to the Networking Research at UNC page.