A Framework for Using Neural Networks for Web Proxy Cache Replacement

Web traffic has continued to increase at a significant rate and is resulting in considerable strain on web servers and bandwidth providers, such as Internet Service Providers (ISPs). Proxy caches are often used to reduce this burden. In this research, a novel approach to web proxy cache replacement is developed. Unlike previous approaches, this research utilizes neural networks for replacement decisions. The neural network is given information about objects in the cache known to be important in web proxy caching as input and the network's output is used to guide cache object replacement.