Design and Analysis of a Fault-Tolerant Mechanism for a Server-Less Video-On-Demand System
ICPADS '02 Proceedings of the 9th International Conference on Parallel and Distributed Systems
Decentralized Load Balancing on Unstructured Peer-2-Peer Computing Grids
NCA '06 Proceedings of the Fifth IEEE International Symposium on Network Computing and Applications
Can internet video-on-demand be profitable?
Proceedings of the 2007 conference on Applications, technologies, architectures, and protocols for computer communications
TRIBLER: a social-based peer-to-peer system: Research Articles
Concurrency and Computation: Practice & Experience - Recent Advances in Peer-to-Peer Systems and Security (P2P 2006)
On ISP-friendly rate allocation for peer-assisted VoD
MM '08 Proceedings of the 16th ACM international conference on Multimedia
Scaling peer-to-peer video-on-demand systems using helpers
ICIP'09 Proceedings of the 16th IEEE international conference on Image processing
A grouping algorithm of helpers in peer-to-peer video-an-demand systems
ICACT'10 Proceedings of the 12th international conference on Advanced communication technology
Hi-index | 0.00 |
We propose a decentralized Peer-to-Peer (P2P) Videoon-Demand (VoD) system. The traditional data center architecture is eliminated and is replaced by a large set of distributed, dynamic and individually unreliable helpers. The system leverages the strength of numbers to effect reliable cooperative content distribution, removing the drawbacks of conventional data center architectures including complexity of maintenance, high power consumption and lack of scalability. In the proposed VoD system, users and helper "servelets" cooperate in a P2P manner to deliver the video stream. Helpers are preloaded with only a small fraction of parity coded video data packets, and form into swarms each serving partial video content. The total number of helpers is optimized to guarantee high quality of service. In cases of helper churn, the helper network is also able to regenerate itself by users and helpers working cooperatively to repair the lost data, which yields a highly reliable system. Analysis and simulation results corroborate the feasibility and effectiveness of the proposed architecture.