The long-term success of the World Wide Web depends on fast response time. People use the Web to access information from remote sites, but do not like to wait long for their results. The rapid growth in the amount of information and the number of users has lead to difficulty in providing effective response time for the web users and this increased web latency; resulting in decreased web performance. Although several proposals have been made for reducing this latency, like it can be improved by caching, the benefit of using it is rather limited owing to filling the cache with documents without any prior knowledge.Predictive caching becomes an attractive solution wherein the forthcoming page likely to be requested soon are predicted based on user access logs information and pre-fetched ,while the user is browsing the current display pages. As web page prediction gained its importance, This paper proposes a bracing approach for increasing web server performance by analyzing user behavior, in this pre-fetching and prediction is done by pre-processing the user access log and integrating the three techniques i.e. Clustering, Markov model and association rules which achieves better web page access prediction accuracy;This work also overcomes the limitation of path completion i.e. by extracting web site structure paths are completed,which helps in better prediction, decreasing access time of user and improving web performance.

No abstract

The MANET faces major challenges in design of routing protocol due to their highly dynamic topology, poses by mobility. The position based routing algorithm that does not depend on topological information is more suitable than other routing protocol for MANET. In previous position based routing protocol does not consider the traffic patterns of higher layer. In fact, the result of these algorithms and test of connectivity in the network layer level and the used to analyze position based algorithm that do not reflect the true behavior of MANET. In our proposed position based routing protocol, we have made modifications in LAR1 algorithm by changing the request zone region computations. The results of this algorithm are compared with the existing LAR1. The performance of the CHMFR protocols is measured in terms of throughput, success rate and power consumption for varying network size which is efficient as compared to LAR1. We have simulated CHMFR protocol using GloMoSim simulator.

No abstract