Dynamic Multicast Tree to Support Time-Shifting in Real-Time Streaming

Weihui,; Xu, Jikun; Cong, Qing

doi:10.1109/icicci.2010.30

Cited by 1 publication

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PPB, Peer-toPeer Batching, combines peer-to-peer and IP multicast by using IP multicast for batching while a peer-to-peer network is used to provide the initial portion of video to new peers [20]. In [21], the authors propose a tree-based system that constructs dynamic multicast trees, called Jtree. The J-tree structure explicitly distinguishes the live multicast tree and playback trees that are implicitly built in our system.…”

Section: Introductionmentioning

confidence: 99%

Time-Shifted Streaming in a Tree-Based Peer-to-Peer System

Noh¹,

Girod²

2012

JCM

View full text Add to dashboard Cite

Abstract-We propose a peer-to-peer (P2P) streaming system that multicasts live video as well as provides the live video as Video-on-Demand (VoD) during the live session. The proposed system allows users to individually pause and resume a live video stream, and to view video that was streamed before the particular user joined the session. Since live video content naturally results in many users watching it online at the same time, the P2P concept is leveraged to reduce server load.We extend our Stanford Peer-to-Peer Multicast (SPPM) protocol, originally designed for live video multicast, to support playback control. To achieve this in a P2P fashion, peers store received video packets in their local buffer and forward them at a later time when requested by other peers, which is called time-shifted streaming. To understand interaction between live and time-shifted traffic, we analyze video availability of the proposed system, which refers to how many peers possess video contents of a particular position. Motivated by the analysis, we propose fast prefetching and a parent selection scheme suitable for fast prefetching in order to further reduce server load by disseminating video quickly. Fast prefetching is possible because the relaxed transfer deadline of time-shifted streams and peers' extra uplink bandwidths are exploited. With simulations and mathematical analysis, it is shown that fast prefetching not only alleviates the requirement of server bandwidth by expediting video dissemination, but also mitigates video disruption due to peer churn.

show abstract

Section: Introductionmentioning

confidence: 99%