Web caching framework: analytical models and beyond

Abstract

Many Web caching algorithms have been proposed in recent years. However the lack of analytical support and systematic evaluation environment significantly affect the applicability of these algorithms. We introduce a framework within which Web caching algorithms can be consistently analyzed and empirically examined. The framework consists of two complementary parts. The statistical model and the simulation environment. The analytical model covers both the Web trace characteristics and the caching algorithm behaviors. The simulation system, referred to as WebCASE (Web Caching Algorithm Simulation Environment), consists of an extensible simulation core and a front end graphical interface showing the running algorithm behaviors. By using this framework, we are able to better understand the performance discrepancies exhibited by different algorithms and develop more efficient new algorithms. These new algorithms take into consideration practical issues and make noticeable performance improvements over existing algorithms.