Document Type





parallel input-output, media-on-demand server, striping, real-time data retrieved, data access patterns




Computer Sciences


One of the key components of a multi-user multimedia-on-demand system is the data server. Digitalization of traditionally analog data such as video and audio, and the feasibility of obtaining network bandwidths above the gigabit-per-second range are two important advances that have made possible the realization, in the near future, of interactive distributed multimedia systems. Secondary-to-main memory I/O technology has not kept pace with advances in networking, main memory and CPU processing power. Consequently, the performance of the server has a direct bearing on the overall performance of such a system. In this paper we present a high-performance solution to the I/O retrieval problem in a distributed multimedia system. We develop a model for the architecture of a server for such a system. Parallelism of data retrieval is achieved by striping the data across multiple disks. We identify the design parameters that affect the throughput of the server. We have implemented our model on the Intel Paragon parallel computer. We have performed an extensive performance evaluation of how the parameters identified affect the data retrieval efficiency of the server. The results of component-wise instrumentation of the server operation are presented and analyzed. The performance of any server ultimately depends on the data access patterns. Two modifications of the basic retrieval algorithm that exploit data access patterns in order to improve system throughput and response time are presented. Based on our experiments, a dynamic admission control policy that takes server workload into account is proposed.