Neuro-fuzzy system in web client-side caching

Abstract

Web caching is a well-known strategy for improving performance of Web-based system by keeping web objects that are likely to be used in the near future closer to the client. Although most researchers focused on designing efficient caching with proxy and origin servers, the potential gain of exploiting client-side caching based on neuro-fuzzy system is not yet being investigated. Hence, this paper proposes a splitting web client-side cache to two caches, short-term cache and longterm cache. Initially, a web object is stored in short-term cache. The web objects that are visited more than the pre-specified threshold value will be moved to long-term cache and other objects in short-term cache are removed with time. Thus, we ensure that the preferred web objects are cached in long-term cache for longer time. In this study, neuro-fuzzy is employed to determine which web objects should be removed in order to create more spaces for the new web objects. By implementing this mechanism, the cache space is used properly. Experimental results have shown that the proposed approach has better performance compared to the most common caching policies and has improved the performance of client-side caching substantially.