Title :
Performance evaluation of cooperative peer selection methods for P2P Video-on-Demand
Author :
Kawarasaki, Masatoshi ; Suzuki, Kei
Author_Institution :
Grad. Sch. of Libr., Inf. & Media, Studies, Univ. of Tsukaba, Tsukuba, Japan
Abstract :
Recently, the video delivery service is spreading along with the penetration of broadband access to residential users. Although content delivery network (CDN) system has been deployed for video delivery, it´s costly because equipment reinforcement is needed as a number of client increases. To settle this problem, use of peer-to-peer (P2P) system has been considered, namely P2P live-streaming and P2P-VoD (Video on Demand). P2P-VoD is more difficult than P2P live-streaming as each peer is watching different part of the video. P2P-VoD could be achieved based on P2P live-streaming or on BitTorrent. However, both include respective issues. This paper focuses on BitTorrent-based VoD and proposes methods to overcome its issues such as server dependency, network efficiency and fairness. In our methods, the tracker selects cooperating peers by being aware of their available pieces or line speeds, and the cooperating peers exchange pieces by adjusting the number of unchoke peers according to the occurrence rate of video request. A simulation study is performed to show the effectiveness of the proposed methods.
Keywords :
peer-to-peer computing; performance evaluation; video on demand; video streaming; P2P system; content delivery network; cooperative peer selection; live streaming; peer-to-peer system; performance evaluation; video delivery service; video-on-demand; Inductors; Internet; Libraries; Network servers; Peer to peer computing; Protocols; Streaming media; Tree data structures; Video on demand; Watches; BitTorrent; fairness; load distribution; peer-to-peer (P2P); video on demand (VoD);
Conference_Titel :
Ultra Modern Telecommunications & Workshops, 2009. ICUMT '09. International Conference on
Conference_Location :
St. Petersburg
Print_ISBN :
978-1-4244-3942-3
Electronic_ISBN :
978-1-4244-3941-6
DOI :
10.1109/ICUMT.2009.5345561