An Efficient Approach For Optimal Prefetching To Reduce Web Access Latency.

ABSTRACT The exponential growth and popularity of WWW increases the amount of traffic which results in major congestion problems over the available bandwidth for the retrieval of data. This results in the increase of user perceived latency. Prefetching of web pages is a potential area that can significantly reduce the web access latency. It refers to the mechanism of deducing the forthcoming page accesses of a client. Prefetching reduces the users perceived latency but on the contrary it increases the traffic that may result in further congestion. So the major concern of the prefetching is to device an algorithm that could efficiently and optimally prefetch the pages so that the traffic load is minimized. In this dissertation an optimal prefetching algorithm is proposed which gives the optimal number of web documents to be prefetched to reduce latency. The algorithm is based on the current content of the web documents so there is no requirement of maintaining past history of the users and is also beneficial for first retrieval of access of web resources.