A case for end system multicast (keynote address)
Proceedings of the 2000 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Scalable application layer multicast
Proceedings of the 2002 conference on Applications, technologies, architectures, and protocols for computer communications
Epidemic live streaming: optimal performance trade-offs
SIGMETRICS '08 Proceedings of the 2008 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Is Random Scheduling Sufficient in P2P Video Streaming?
ICDCS '08 Proceedings of the 2008 The 28th International Conference on Distributed Computing Systems
PRIME: peer-to-peer receiver-driven mesh-based streaming
IEEE/ACM Transactions on Networking (TON)
A Measurement Study of a Large-Scale P2P IPTV System
IEEE Transactions on Multimedia
Inferring Network-Wide Quality in P2P Live Streaming Systems
IEEE Journal on Selected Areas in Communications
R2: Random Push with Random Network Coding in Live Peer-to-Peer Streaming
IEEE Journal on Selected Areas in Communications
Understanding the Power of Pull-Based Streaming Protocol: Can We Do Better?
IEEE Journal on Selected Areas in Communications
| Hi-index | 0.00 |
P2P based live streaming has been gaining popularity. The new generation P2P live streaming systems not only attract a large number of viewers, but also support better video quality by streaming the content at higher bit-rate. In this paper, we propose a novel P2P streaming framework, called Hierarchically Clustered P2P Video Streaming, or HCPS, that can support the streaming rate approaching the optimal upper bound while accommodating large viewer population. The scalability comes with the hierarchical overlay architecture by grouping peers into clusters and forming a hierarchy among them. Peers are assigned to appropriate cluster so as to balance the bandwidth resources across clusters and maximize the supportable streaming rate. Furthermore, individual peers perform distributed queue-based scheduling algorithms to determine how to retrieve data chunks from source and neighboring peers, and how to utilize its uplink bandwidth to serve data chunks to other peers. We show that queue-based scheduling algorithms allow to fully utilize peers' uplink bandwidths, and HCPS supports the streaming rate close to the optimum in practical network environment. The prototype of HCPS is implemented and various design issues/tradeoffs are investigated. Experiments over the PlanetLab further demonstrate the effectiveness of HCPS design.