Performance analysis of data block synchronization mechanism in coolstreaming

Abstract

Coolstreaming is a Peer-to-Peer (P2P) based video streaming system in which a single video stream is decomposed into multiple sub-streams and a client-peer node receives the sub-streams from multiple parent-peer nodes, combining them into the original video stream. The buffer of the client-peer node is composed of a synchronization buffer and a cache buffer. Data blocks are stored in the synchronization buffer in a sub-stream basis, and then forwarded into the cache buffer according to their sequence numbers. Here, data-block synchronization is important to guarantee video quality. In this paper, we consider the performance of data-block synchronization scheme with which data blocks are simultaneously forwarded just after all the data blocks composing a macro data block arrive at the synchronization buffer. We model the synchronization buffer as a multiple-buffer queueing system with homogeneous Poisson arrival processes, deriving the mean forwarding interval. We also consider the frame loss probability for multiple-path video streaming, investigating how the number of sub-streams decreases the frame loss probability. Numerical examples show that increasing the number of sub-streams makes the average forwarding interval large, while the frame loss probability at the bottleneck router is improved. It is also shown that increasing the synchronization buffer decreases the average forwarding interval.