Adaptive fuzzy caching algorithm for proxy servers

The problem of increasing the efficiency of cache systems using artificial intelligence techniques is considered. The formalized statement of raising the cache effectiveness on the basis of the set-theoretic mathematical tools is described. The fuzzy logic application technology for determining the victim which implements the calculation of cache values for object in cache memory is considered. It is first offered to use spatial locality of the web-resource as one of its basic characteristics applying to the input of the fuzzy logic system. The definition of this characteristic is based on the uniform resource identifier URI. The technique for the adaptation of fuzzy systems through the synthesis of the linguistic model using the method of fuzzy clustering is defined. Besides, the experimental study of the algorithm developed on the simulation stand for traces obtained through the proxy cache Squid is resulted.