Optimizing the quality of scalable video streams on P2P networks

Raj Kumar Rajendran, Dan Rubenstein
2004 Performance Evaluation Review  
The volume of multimedia data, including video, served through Peer-to-Peer (P2P) networks is growing rapidly. Unfortunately, high bandwidth transfer rates are rarely available to P2P clients on a consistent basis. In addition, the rates are more variable and less predictable than in traditional client-server environments, making it difficult to use P2P networks to stream video for on-line viewing rather than for delayed playback. In this paper, we develop and evaluate on-line algorithms that
more » ... ordinate the pre-fetching of scalably-coded variable bitrate video. These algorithms are ideal for P2P environments in that they require no knowledge of the future variability or availability of bandwidth, yet produce a playback whose average rate and variability are comparable to the best off-line prefetching algorithms that have total future knowledge. To show this, we develop an off-line algorithm that provably optimizes quality and variability metrics. Using simulations based on actual P2P traces, we compare our on-line algorithms to the optimal off-line algorithm and find that our novel on-line algorithms exhibit near-optimal performance and significantly outperform more traditional pre-fetching methods. ¢ layers produces the best quality. The reader is referred to [4] and [9] for more details on MPEG-4 and FGS coding. III. PROBLEM FORMULATION In this section we formulate the layer pre-fetching problem for which we will design solution algorithms. We begin by describing how the video segments, or chunks, the atoms of the video that will be pre-fetched, are constructed. We then formally define the metrics of interest in the context of these chunks, such that our optimization problems to minimize these metrics are well-posed. A. The Model Our model is discrete. We first partition the video along its running time axis into © meta-chunks, which are data blocks of a fixed and pre-determined byte size (the © t h chunk may of course be smaller) 1 . These meta-chunks are numbered 1 through © in the order in which they are played out to view the video. We let
doi:10.1145/1012888.1005735 fatcat:4ajuqeehhfa5rprp7kpj5gctvu