Probabilistic methods for web caching

We introduce and analyze new randomized policies for the management of web caches. The proposed policies are fully scalable, that is, handling a hit or an eviction requires only O(1) time. Our analysis is probabilistic, in nature, and based on an extended version of the IR model. The extension is needed in order to deal with the varying-cost and varying-size features of web documents. Under this assumption, we derive closed-form expressions for the stationary probabilities of nding each possible arrangement of the documents within the cache. Our analysis shows that the performance of the proposed algorithms are close to that of the optimal o -line algorithm. Using simulations and real traces, we also show that the new algorithms perform at least as well as existing algorithms of higher complexity. Variations on the algorithms, aimed at increasing their responsiveness to non-stationary trends, are also investigated. This research was partially supported by NSF grant ANI-9872764. Corresponding author. The research of this author was performed at the University of California Berkeley, under a fellowship for prospective researchers from the Swiss National Science Foundation.