Design of a Peer-to-Peer Video-on-Demand System with the Consideration of Fault Exception

Early departure of customers during video playback brings a great impact on the design of a peer-to-peer (P2P) video-on-demand (VoD) system, especially on the system bandwidth requirement and the need for a well-defined fault exception mechanism. In this paper, we develop an analytical model to evaluate how the system parameters such as bandwidth requirement and batching time in a P2P batching system are affected when early departure behavior is taken into account. In addition, a fault exception mechanism is also proposed. Computer simulations are performed to verify the correctness of the model. The results show that the system resources can be utilized more effectively when the customer departure behavior is captured into the system.

[1]  Jian Feng,et al.  Batching policy for video-on-demand in multicast environment , 2000 .

[2]  Philip S. Yu,et al.  Exploring wait tolerance in effective batching for video-on-demand scheduling , 1998, Multimedia Systems.

[3]  Avideh Zakhor,et al.  Multiple sender distributed video streaming , 2004, IEEE Transactions on Multimedia.

[4]  K.H. Ho,et al.  Enhanced Peer-to-Peer Batching Policy for Video-on-Demand System , 2006, 2006 International Symposium on Communications and Information Technologies.

[5]  Te-Chou Su,et al.  Optimal chaining scheme for video-on-demand applications on collaborative networks , 2005, IEEE Transactions on Multimedia.

[6]  Tomasz Imielinski,et al.  Metropolitan area video-on-demand service using pyramid broadcasting , 1996, Multimedia Systems.

[7]  Donald F. Towsley,et al.  A peer-to-peer on-demand streaming service and its performance evaluation , 2003, 2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698).

[8]  Li-Ming Tseng,et al.  Harmonic broadcasting for video-on-demand service , 1997, IEEE Trans. Broadcast..

[9]  Bharat K. Bhargava,et al.  On peer-to-peer media streaming , 2002, Proceedings 22nd International Conference on Distributed Computing Systems.