TRACE-DRIVEN SIMULATION OF DOCUMENT CACHING STRATEGIES FOR INTERNET WEB-SERVERS

Given the continued growth of the World-Wide Web, performance of Web s ervers is becoming increasingly important. File caching can be used to reduce the time that it takes a Web server to respond to client reque sts, by storing the most popular files in the main memory of the Web s erver, and by reducing the volume of data that must be transferred bet ween secondary storage and the Web server. In this paper, we use trace -driven simulation to evaluate the effects of various replacement, thr eshold, and partitioning policies on the performance of a Web server. The workload traces for the simulations come from Web server access lo gs, from six different Internet Web servers. The traces represent thre e different orders of magnitude in server activity and two different o rders of magnitude in time duration. The results from our simulation s tudy show that frequency-based caching strategies, using a variation o f the Least Frequently Used (LFU) replacement policy, perform the best for the Web server workload traces considered. Thresholding policies and cache partitioning policies for Internet Web servers do not appear to be effective.