Effective Memory Use in a Media Server

Abstract

A number of techniques have been developed for maximizing disk bandwidth utilization in media servers, including disk arm scheduling and data placement ones. Instead, in this paper we focus on how to efficiently utilize the available memo ry. We present techniques for best memory use under different disk policies, and derive precise formulas for computing throughput and memory use. We show that with proper memory use, maximizing disk bandwidth utilization does not necessarily lead to optimal throughput. In addition, we study the worst-case initial latency of presentations, a factor that may be important in interactive systems. We present a technique for significantly reducing this initial latency under some disk scheduling schemes.