Efficient load reduction and congession control in Internet through multilevel Border Gateway Proxy Caching

The enormous popularity of World Wide Web and its application of searching have caused a tremendous increase in network traffic and congestion in network. This project solves this problem by using the techniques of web caching and web prefetching implemented in our proxy system. Web caching has introduced as an effective solution to the problems of traffic congestion and effective bandwidth utilization over the Web. The efficiency of this cache depends on the chosen cache update scheme, the type and number of levels of cache used in implementation. Instead of relying on the traditional single level cache implementation in proxy it is replaced by the most efficient and beneficial multilevel cache for much more tuned performance and access. Each level of cache is to be evicted and updated regularly which is done by use of History based LRU and LFU as a best strategic policy for cache eviction and updation. This client side proxy server provides the cached documents to the requested clients upon request. Instead of implementing this for single LAN we can implement this idea for cache sharing among multiple LAN´S so that more clients get benefited. This is done by setting a dedicated individual proxy server near the border gateway router of each LAN. Each border proxy system implements the above discussed caching mechanisms and handles its own client´s of the LAN and they communicate with other border gateway proxy systems through Inter Autonomous system routing mechanism. As a result the cached data in one LAN can be shared by the other LAN´s through this gateway proxy which can save the congession produced by the respective LAN´S. The above all concepts are implemented as a single proposed algorithm namely Border Gateway Proxy Caching algorithm (BGPC algorithm). Apart from the above functionality it also does some additional functionality like log maintenance, request handling and cache-crash management.