Performance and Quality-of-Service Analysis of a Live P2P Video Multicast Session on the Internet

Abstract

Abstract-We evaluate the performance of a large-scale live P2P video multicast session comprising more than 120, 000 peers on the Internet. Our analysis highlights P2P video multicast characteristics such as high bandwidth requirements, high peer churn, low peer persistence in the P2P multicast system, significant variance in the media stream quality delivered to peers, relatively large channel start times, and flash crowd effects of popular video content. Our analysis also indicates that peers are widely spread across the IP address space, spanning dozens of countries and hundreds of ISPs and Internet ASes. As part of the P2P multicast evaluation several QoS measures such as fraction of stream blocks correctly received, number of consecutive stream blocks lost, and channel startup time across peers. We correlate the observed quality with the underlying network and with peer behavior, suggesting several avenues for optimization and research in P2P video multicast systems.

Free-Riding, Fairness, and Firewalls in P2P File-Sharing

“…The percentages of firewalled peers we measured are consistent with previous measurements [2,5,6,17,[21][22][23], but these vary widely in their results (between 25% and 93%). Firewall puncturing and NAT traversal techniques have been researched [4,[7][8][9]19], and proven to be quite effective for the considered sets of firewalls and NATs.…”

“…The reason for the difference between these communities is beyond our abilities to measure, but the difference does correlate with our theory. Even though the measurements of both communities were taken 20 months apart from each other, measurements by others do not suggest a decrease in the fraction of firewalled peers over time [2,5,6,17,[21][22][23]. We believe this difference to be due to the policy of banning peers with low sharing ratios in closed communities such as TVTor- rents.…”

Service quality in P2P streaming systems

“…Finally, as the quality is estimated using the buffer content of peers it may not correlate well with perceived quality experienced by users. Note that similarly to [12,13,19] Hei et al also report that some peers experienced inadequate service quality.…”

“…In [12], the authors extended their work to the case of operational P2P steaming system. In that study they set-up an undisclosed P2P system to broadcast a very popular sporting event that attracted more than 120000 viewers across the world.…”

“…If some of those are congested the resulting performance may degrade substantially. For example, in [12] long distance between communicating peers was identified as the main reason for quality degradation. To tackle this problem CDN functionality can be introduced to the system.…”