A measurement of a large-scale peer-to-peer live video streaming system

Peer-to-Peer (P2P) technologies have found much success in applications like file distributions, and its adoption in live video streaming has recently attracted significant attentions. With the emerge of commercial P2P streaming systems that are orders of magnitude larger than the academic systems, understanding its basic principles and limitations are important in the design of future systems. Coolstreaming represented one of the earliest large-scale live streaming trials in the Internet. In this paper, we discuss the fundamental design principles and examine the system dynamics. By leveraging the recent results obtained from live event broadcast, we develop some basis to demonstrate that a random partnership selection has the potential to scale. Specifically, first, we examine the overlay topology and how it converges. Second, using a combination of real traces and analysis, we quantitatively provide insights on how buffering technique resolves the problems associated with dynamics and heterogeneity. Third, we discuss the main limitations and the implications on the scalability.