Query Adaptation Techniques in Temporal-DHT for P2P Media Streaming Applications

Peer-to-Peer (P2P)-based approach for on-demand video streaming systems (P2P-VoD) characterized by asynchronous user-interactivity has proven to be practical and effective in recent years with real-world Internet-scale deployment (Huang, Li, & and Ross, 2007). Current state-of-art P2P-VoD systems employ tracker server for discovering content suppliers which poses scalability and bottleneck issues. Temporal-DHT is a structured P2P based approach which can efficiently accommodate the large number of update operations with the continuous change of user’s playing position and supporting asynchronous jumps (Bhattacharya, Yang, & Zhang, Temporal DHT and its Application in P2P-VoD Systems., 2010). We propose different query adaptation strategies based upon content popularity distributions and shortage bandwidth ratios which are proved to be effective in improving the performance of P2P streaming system by deriving certain optimized solutions. We formulate valuable optimization problems in the context of a P2P-VoD system such as minimization of query search cost, server bandwidth consumption, and a joint cost-load framework. We provide optimized solutions that achieve the best result for the above mentioned optimization objectives. We show extensive simulation studies under various scenarios of search cost, streaming quality, and other associated factors in a dynamic network environment where users are free to asynchronously join/leave the system.