Dynamic optimization of readsize in hypermedia servers

Abstract

While disk scheduling techniques for continuous media (CM) servers are mostly designed to maximize the number of concurrent CM data streams, considerably less attention has been paid to discrete media (DM) data throughput. DM data throughput is vital to systems that need to support the heterogeneity and variety of data found in interactive hypermedia and digital library applications. Therefore, a server's ability in delivering a high DM data throughput without degrading its CM data throughput deserves more attention. A new technique is introduced to optimize the use of disk bandwidth at run-time and redirect the saved bandwidth to service DM requests. Based on a new cost model of disk access incurred in CM service, we formulate strategies to moderate the size of disk reads at run-time. Through experimental evaluations, these strategies are found to improve the efficiency of disk accesses and the DM data throughput of a hypermedia server without jeopardizing the throughput and quality of CM service.