A scalable based multicast model for P2P Conferencing applications

Multicast conferencing is a rapidly-growing area of Internet use. Audio, video and other media such as shared whiteboard data can be distributed efficiently between groups of conference participants using multicast algorithms that minimize the amount of traffic sent over the network. This is far more effective than systems that maintain a separate link between each participant. On the other hand Peer-to-Peer (P2P) model is inherently characterized by high scalability, robustness and fault tolerance. With its decentralized and distributed architecture, a P2P network is somehow able to self organized dynamically. Peer-to-Peer model or architecture is well adapted to conferencing applications, effectively it can greatly benefit from P2P attributes such as: flexibility, scalability and robustness, particularly in critical environments such as: mobile networks. In this paper we propose a novel and scalable approach for P2P conferencing. This model combines a call control and signalling protocol (SIP) with a ¿P2P¿ protocol (Chord) for maintaining (dynamically) a well stabilized and optimized architecture topology. This model is also based on an application layer multicast mechanism. Performance evaluation shows that our proposed approach benefits from SIP protocol flexibility, with the robustness and scalability of chord protocol. The use of a multicast mechanism optimizes the overall traffic flow (control and media) and transmission efficiency.