On the Sensitivity of Web Proxy Cache Performance to Workload Characteristics

This paper describes the design and use of a synthetic Web proxy workload generator (ProWGen) to investigate the sensitivity of proxy cache replacement policies to selected Web workload characteristics. Tracedriven simulations with synthetic workloads from ProWGen show the relative sensitivity of three popular cache replacement algorithms -- LRU, LFUAging, and GD-Size -- to Zipf slope, temporal locality, and correlation (if any) between file size and popularity, and the relative insensitivity of these algorithms to one-timers and heavy tail index. Performance differences between the three policies are also highlighted