Use of analytical performance models for system sizing and resource allocation in interactive video-on-demand systems employing data sharing techniques

Abstract

AbstractÐIn designing cost-effective video-on-demand (VOD) servers, efficient resource management and proper system sizing are of great importance. In addition to large storage and I/O bandwidth requirements, support of interactive VCR functionality imposes additional resource requirements on the VOD system in terms of storage space, as well as disk and network bandwidth. Previous works have used ªdata sharingº techniques (such as batching, buffering, and adaptive piggybacking) to reduce the I/O demand on the storage server. However, such data sharing techniques complicate the provision of VCR functions and diminish the amount of benefit that can be obtained from data sharing techniques. The main contribution of this paper is a simple, yet powerful, analytical modeling approach which allows for analysis, system sizing, resource allocation, and parameter setting for a fairly general class of ªdata sharingº techniques which are used in conjunction with the providing of VCR-type functionality. Using this mathematical model, we can determine the proper amount of resources to be allocated for normal playback as well as for service of VCR functionality requests while satisfying predefined system performance requirements. To illustrate the usefulness of our model, we focus on a specific data sharing scheme which combines the use of batching, buffering, and adaptive piggybacking, as well as allows for the use of VCR functions. We show how to utilize this mathematical model for system sizing and resource allocation purposesÐthat is, how to distribute the available resources between the service of normal playback and VCR functionality requests under various workloads and resource price ratios, so as to obtain the lowest system cost.

Pinned demand paging based on the access frequency of video files in video servers

“…Furthermore, Leung et al (2002) propose a simple, yet powerful, analytical modeling approach which allows for analysis, system sizing, resource allocation, and parameter setting for a fairly general class of data sharing techniques (i.e., buffering and batching) which are used in conjunction with the providing of VCR (video cassette recorder)-type functionality.…”

Performance evaluation for VBR Continuous Media File Server admission control

“…This has been analyzed and simulated by several researchers [15,27,28,30]. The approaches range from batching to patching.…”

“…One other approach is adaptive piggybacking [29], which alters the frame display rate at the client to enable merging of requests. Further analysis on sharing [30] examines bandwidth savings and system capacity planning issues. Some of these approaches replicate video files across striped storage media for hight performance in the normal case, while providing fault tolerance and graceful degradation in the case of failures.…”

Providing VCR Functionality in VOD Servers

“…Researchers have investigated the impact of VCR operations on the performance of batching [14,15,16,17,18,28], piggybacking [14,17,18,28], patching [13,19,26,28], chaining [25,28,29], and hierarchical-stream merging [24,26,27]. Some researchers offer solutions to support VCR functionality in VOD servers with resource-sharing technology.…”

“…• Contingency Pool: References [14,15,16,17,18,19] take the contingency-pool approach, which can be used with any resource-sharing technology. The idea is to reserve a pool of channels to support potential VCR operations.…”