Scalable Peer-to-Peer Streaming Protocol for Real Time Data
- 詳細技術說明
- This scalable Peer-to-Peer protocol uses an elegant network formation and maintenance algorithm that minimizes the issues of peer churn, bum peers, and delay, providing for maximum theoretical scaling.
- *Abstract
-
Current direct real time streaming implementations on the internet are limited by bandwidth consumption. Peer-to-Peer streaming reduces server bandwidth requirements, but it is vulnerable to delay and reliability issues that multiply as the network scales in size. Current Peer-to-Peer streaming implementations do not scale well and are not robust.
This scalable Peer-to-Peer protocol uses an elegant network formation and maintenance algorithm that minimizes the issues of peer churn, bum peers, and delay, providing for maximum theoretical scaling.
DESCRIPTION/DETAILS
This invention is an elegant but straightforward network management algorithm for the formation of efficient Peer-to-Peer streaming networks that scale with minimal delay.
It is well known that scaling of server-client direct streaming is hindered by bandwidth requirements. Each client that is added requires a full addition of bandwidth communication to the server. N clients results in N times the data streamed worth of bandwidth consumption. This is a costly approach to live data distribution.
In the same way that Peer-to-Peer networks have revolutionized the bandwidth consumption model for file sharing, Peer-to-Peer distribution models can alleviate the bandwidth requirements of the source of a broadcast. However, live data presents unique challenges. Delays in delivery can result from the processing and passing of the stream by a number of upstream peers (latency). Low-bandwidth upstream peers can also throttle the total data pass-through (bum clients). The constant leaving and joining of new clients can cause missing links in the chain of broadcast (peer churn).
This Peer-to-Peer algorithm addresses each of these issues, reducing latency to its theoretical minimum values, providing multiple but efficient paths of data flow to smooth differences in bandwidth and procession power between peers, and handling peer churn through self-regulating broadcast networks that seamlessly insinuate new peers and bridge the gaps left by departed ones.
The result is the formation of a robust Peer-to-Peer network that is uniquely scalable to meet the needs of live data distribution. It can scale the layers of its network to consume available bandwidth to enhance reliability and quality, and it can scale to reach larger audiences with minimized latency and performance degradation.
APPLICATIONS
- Live internet broadcast of video or timely data to large audiences.
- Internet radio and television distribution, video game broadcast, personal broadcasting through social media.
BENEFITS
- P2P broadcasts do not require server and bandwidth resources, empowering all internet users to broadcast to large audiences.
- Scaling algorithm allows maximum broadcast scaling with minimal drawbacks.
For more information about this technology, please contact the University of Illinois at Urbana-Champaign Office of Technology Management at otm@illinois.edu.
- 國家/地區
- 美國
