Consumption-based buffer management for maximizing system throughput of a multimedia system

In a multimedia server, multiple media streams are generally serviced in a cyclic fashion. Due to non-uniform playback rates and asynchronous arrivals of queries, there tends to be spare disk bandwidth in each service cycle. In this paper, we study the issue of dynamically utilizing the spare disk bandwidth and buffer to maximize the system throughput of a multimedia server. We introduce the concept of minimizing buffer consumption to select an appropriate media stream to use the spare disk bandwidth. The buffer consumption measures both the amount of buffer and the amount of time such a buffer is occupied (i.e. the space-time product). Different alternatives to utilizing spare disk bandwidth are examined, including rate-adjustable retrievals of an already-activated stream and prefetching the next waiting stream. Simulations are conducted to evaluate and compare different cases. The recruits show that (1) minimizing buffer consumption is a good criterion for maximizing the system throughput, and (2) in general, prefetching a waiting stream incurs more buffer consumption, and thus is less effective than rate-adjustable retrievals.