Sources and characteristics of Web temporal locality
S. Jin, Azer Bestavros · 2002
Temporal locality of reference in Web request streams emerges from two distinct phenomena: the long-term popularity of Web documents and the short-term temporal correlations of references. We show that the commonly-used distribution of inter-request times is predominantly determined by the power law governing the long-term popularity of documents. This inherent relationship tends to disguise the existence of short-term temporal correlations. We propose a new and robust metric that enables accurate characterization of that aspect of temporal locality. Using this metric, we characterize the locality of reference in a number of representative proxy cache traces. Our findings show that there are measurable differences between the degrees (and sources) of temporal locality across these traces.