Comparison of piece-picking algorithms for layered video content in peer-to-peer networks

The distribution of video content over P2P systems has become a popular and cost-effective option in recent years due to the increasing online availability of content. Additionally, the diversity of end-user terminals used for consuming content demands the provisioning of content in different qualities. Both problems are addressed by distributing layered video content over P2P networks. A piece-picking algorithm for layered video content needs to ensure that the pieces are received in time as well as that the best possible quality that can be processed by the end-user terminal and downloaded under the given network conditions is provided. In this paper, we describe our algorithm for piece-picking of layered video content in Bittorrent-based P2P systems and compare it to other existing piece-picking algorithms. The evaluation presented in this paper shows that our algorithm can very well compete with previously published algorithms.