Biased Replacement Policies for Web Caches: Di erential Quality-of-Service and Aggregate User Value

Disk space in shared Web caches can be diverted to serve some system users at the expense of others. Cache hits reduce server loads, and if servers desire load reduction to di erent degrees, a replacement policy which prioritizes cache space across servers can provide di erential quality-of-service (QoS). We present a simple generalization of leastfrequently-used (LFU) replacement that is sensitive to varying levels of server valuation for cache hits. Through trace-driven simulation we show that under a particular assumption about server valuations our algorithm delivers a reasonable QoS relationship: higher byte hit rates for servers that value hits more. We furthermore adopt the economic perspective that value received by system users is a more appropriate performance metric than hit rate or byte hit rate, and demonstrate that our algorithm delivers higher \social welfare" (aggregate value to servers) than LRU or LFU. Submitted to the Fourth International Web Caching Workshop, San Diego, California, 31 March{2 April 1999. Acceptance noti cation deadline is 15 February 1999. Details available at http://www.ircache.net/Cache/Workshop99/. yContact author. Biased Replacement Policies January 22, 1999 1