Application of Multifractals in WWW Traffic Characterization

Marwan Krunz

University of Arizona


WWW caching is used to improve network latency and bandwidth usage by storing previously requested files in a cache. Ideally, the cache replacement policy should account for the intrinsic characteristics of WWW traffic, which include temporal locality, spatial locality, and popularity. In this talk, we present a stochastic model that can accurately capture these three characteristics. The model can be used to generate synthetic WWW traces and assess WWW cache designs. To capture temporal and spatial localities, we resort to multifractal analysis, which enables us to capture both properties simultaneously. Multifractality is a generalization of monofractality (i.e., self-similar), where now the Hurst parameter is not constant, but varies with scale. We show that the multifractal approach is quite accurate in reproducing the essential characteristics of WWW traffic. Simulation studies of various cache replacement policies driven by several WWW traffic models are conducted and used to support the goodness of the multifractal approach.

[Marwan Krunz]
[University of Arizona]