Characterizing Document Types to Evaluate Web Cache Replacement Policies

In this paper, a study of the performance of six replacement policies has been developed. Three of them are classical caching algorithms (LRU, LFU and LFU-DA) while the other three are caching schemes specifically developed for Web documents (GO-SIZE, GDSF and GD*). This study has been divided by the main content-types observed in Web traffic (application, audio, images, text and video). Using a trace log of a real proxy cache, a characterization of the main properties of the documents has been performed, such as size characteristics, popularity and temporal locality. Finally, a trace driven simulation study of the performance of the replacement policies has been developed for the traffic generated by each considered content-type