Bandwidth Skimming: A Technique for Cost-Effective Video-on-Demand

Authors: Derek Eager, Mary Vernon, John Zahorjan

Complete Citation

DL Eager, MK Vernon, J Zahorjan. Bandwidth Skimming: A Technique for Cost-Effective Video-on-Demand . Proc. of ACM/SPIE Multimedia Computing and Networking 2000
Download the PDF

Abstract

Video-on-demand applications must consider the bandwidth limitations at the server, within the network, and at the client. Recent multicast delivery techniques have addressed the server and network bandwidth bottlenecks. These techniques, however, neglect consideration of client network access bandwidth limitations. Client bandwidth is often insufficient to permit streaming video of the quality that clients would desire, yet these new multicast delivery techniques require clients to be capable of receiving two or more transmission streams simultaneously, each at the playback bit rate. The reduction in the playback bit rate required to use these techniques implies substantially poorer display quality. This paper proposes a new technique for on-demand delivery of streaming media that addresses this problem. The idea is to hold in reserve, or “skim”, a portion of the client reception bandwidth that is sufficiently small that display quality is not impacted significantly, and yet that is nonetheless enough to support substantial reductions in server and network bandwidth through near-optimal hierarchical client stream merging. In this paper we show that this objective is feasible, and we develop practical techniques that achieve it. The results indicate that server and network bandwidth can be reduced to on the order of the logarithm of the number of clients who are viewing the object, using a small “skim” (e.g., 15%) of client reception bandwidth. These low server and network bandwidths are achieved for every media file, while providing immediate service to each client, and without having to pre-load initial portions of the video at each client.

Annotations

The paper discusses a method of performing HMSM while making allowances for client bandwidth limitations. They use 85% of the client bandwidth for the active stream, and use the remaining 15% to perform the multicasting aspect. They argue that using significantly more for the merging streams does not provide benefit relative to the BW given, and via testing 15% shows an optimal balance of bandwidth to performance gain.

This is significant because standard Hierarchical Multicast Stream Merging systems generally use 50% of the BW for the main stream, and 50% for the merge stream. Dropping the merge stream to 15% allows the main stream (which is being played) to be higher quality. In short, for a minor increase in time-to-merge, they are able to support better quality video to the client.

This is interesting to us because it is one of the few papers I have located that discusses client-side BW constraints. However, they differ from us in that they are addressing client-side download speed, and make no mention of the issues involved with clients being requested to upload at significant rates.

-- DavidMoore - 28 Nov 2007