Near video-on-demand systems with combining multicasting and unicasting batching

Multicasting technology has been a practical solution for implementing large scale video-on-demand (VoD) system. The major drawback of true VoD systems is cost due to the dedicated transmission channel for every single customer. The tradeoffs of the near VoD system, however, come to the scheduled multicasting streams and limited video selections. A near VoD architecture combining multicasting and unicasting batching is proposed in this research. The dynamic adjustment of the batching time with virtual community is presented to solve the server loading and bandwidth problem. Our proposed schemes are believed to provide several strategies for the realistic implementation of the VoD system.