Popularity-Based Selective Markov Model

Web prefetching is a promising solution used to reduce user´s latency and improve the QOS. This paper presents a popularity-based selective Markov prefetching model for predicting the forthcoming Web pages. We make use of teh Zipf´s law to model the Web objects´ popularity. An experimental evaluation of the prefetching mechanism is presented using real server logs. Our trace-driven simulation results show that the popularity-based selective. Markov prefetching model can achieve a good hit ratio with reducing the traffic load to some degree.