Performance of P2P live video streaming systems on a controlled test-bed
Proceedings of the 4th International Conference on Testbeds and research infrastructures for the development of networks & communities
Peer assisted video streaming with supply-demand-based cache optimization
IEEE Transactions on Multimedia - Special section on communities and media computing
Detecting malicious nodes in peer-to-peer streaming by peer-based monitoring
ACM Transactions on Multimedia Computing, Communications, and Applications (TOMCCAP)
Optimal bandwidth assignment for multiple description coding in media streaming
CCNC'09 Proceedings of the 6th IEEE Conference on Consumer Communications and Networking Conference
Peer-to-peer video-on-demand with scalable video coding
Computer Communications
A study on upload capacity utilization with minimum delay in peer-to-peer streaming
NEW2AN'11/ruSMART'11 Proceedings of the 11th international conference and 4th international conference on Smart spaces and next generation wired/wireless networking
Proceedings of the 2013 Research in Adaptive and Convergent Systems
A P4P-integrated data-driven P2P system for the live multimedia streaming service
Computer Communications
| Hi-index | 0.25 |
Peer-to-peer (P2P) networks have been adopted for Internet live video-streaming service, and several practical systems have been deployed in past years due to the inherent scalability and ease of deployment. However, most of these systems are commercial and proprietary, and hence little research was done in the area of characterizing practical system performance properties. In this article, we mainly present our experience on a practical P2P-based live video- streaming system called GridMedia, which was employed to broadcast live the Chinese Spring Festival Gala show over the Internet. Benefiting from two sets of flush-crowd traces with about 15,239 and 224,453 concurrent users in a 300 kb/s streaming session in 2005 and 2006, we perform a trace study to understand the service capacity, quality of streaming service, connection heterogeneity, user geographic distribution, and request and online duration characteristics. Our observations shed light on those systems and further improvements in the arena of large-scale live video-streaming service over the Internet.