Adaptive Topology Optimization Base on Bidirectional Peer Selection in Peer-to-Peer Media Streaming

Compared with P2P files sharing, media streaming has more stringent time and bandwidth requirements. Previous work indicates that finding a right peer to query for service will improve performance of a media streaming system. However, most of peer selection algorithms focused on how to choose an appropriate parent node to maximize the requesting peer's profit, while ignoring improving overall service capacity and quality. In this paper, we focus on finding right serving peers and pay more attention to improve the overall performance and quality of service (Qos) of the media streaming system. We propose an adaptive topology optimization policy based on bidirectional peer selection. From the viewpoint of requesting peers, a smoothed value of Round Trip Time (RTT) is used to evaluate the transmission quality; it will help requesting peers to find parents with better link performance. Service capability and contribution of peers are also introduced to help improving overall service capacity and quality. Serving peers perform a client peer adjustment policy to optimize the topology adaptively. The experiment results proved that our policy could improve utilization rate of bandwidth and reduce the depth of multicast tree when compared with delay based and client peer adjustment algorithm.