Most traditional VoD applications rely on a client-server architecture where the video content is stored on a single server or a cluster of servers. The video is then on request streamed or transmitted to the client. These servers usually require tremendous network and processing resources in order to serve multiple clients requests simultaneously. Still, the last mile to the server often becomes the bottleneck of the system because all clients are served by this one central network link. The QoS delivery of video streams cannot be guaranteed in case the bandwidth requirements exceed the resources at disposal. In order to overcome this major drawback network workloads need to be distributed and scaled as it is often achieved in P2P networks. Applying these techniques to VoD applications allows each peer in the network to act as a restricted video-server by offering a low number of selected movies to all other peers. As a consequence the workload is distributed among all participating peers.
The …show more content…
Each peer contains a P2P middleware which ensures the basic communication and interoperability with other peers and in particular with the rendezvous server for content reference exchange. Several subnetworks are interconnected trough routers. The Quality of Service support is provided by the IntServ (Integrated Services) and the MPLS approach. Intserv defines a set of extensions to the traditional best effort model of the Internet with the goal of allowing end-to-end QoS to be provided to applications. The Resource Reservation Protocol (RSVP) represents an implementation of the IntServ idea that determines a fixed path for a video stream in the network and