Boosting Gossip for Live Streaming

Abstract

Abstract-Gossip protocols are considered very effective to disseminate information in a large scale dynamic distributed system. Their inherent simplicity makes them easy to implement and deploy. However, whereas their probabilistic guarantees are often enough to disseminate data in the context of lowbandwidth applications, they typically do not suffice for highbandwidth content dissemination: missing 1% is unacceptable for live streaming.In this paper, we show how the combination of two simple mechanisms copes with this seemingly inherent deficiency of gossip: (i) codec, an erasure coding scheme, and (ii) claim 2°, a content-request scheme that leverages gossip duplication to diversify the retransmission sources of missing information. We show how these mechanisms can effectively complement each other in a new gossip protocol, gossip++, which retains the simplicity of deployment of plain gossip. In a realistic setting with an average bandwidth capability (800 kbps) close to the stream rate (680 kbps) and 1% message loss, plain gossip can provide at most 99% of the stream. Using gossip++, on the other hand, all nodes can view a perfectly clear stream.

CLive: Cloud-assisted P2P live streaming

“…Peer-to-peer (P2P) live streaming is becoming an increasingly popular technology, with a large number of academic [1], [2], [3], [4], [5] and commercial [6], [7] products being designed and deployed.…”

“…CLIVE extends existing mesh-pull P2P overlay networks for video streaming [2], [5], [13], in which each peer in the swarm periodically sends its data availability to other peers, which in turn pull the required chunks of video from the neighbors that have them. The swarm is paired with a CLIVE manager (CM), which participates with other peers in a gossip-based aggregation protocol [14], [15] to find out the current state of the swarm.…”

Resilient gossip algorithms for collecting online management information in exascale clusters

“…Gossip interval: The internal management of colonies requires frequent updates. Examples of gossip intervals are memberlist [10], a library that manages cluster membership and failure detection -200 ms; the gossip++ protocol [11] for boosting live streaming -200 ms; Soltero et al [12] evaluating the use of gossip for power-management in large clusters -1 s; and MOSIX [9], dynamic resource management in clusters -1 s. Local window size: Within colonies, the local window size is only a function of the colony's size and needs. If for example, the colony's management system requires information about all the nodes in the colony of which the average age is below a given value, then Equation (8) provides an appropriate window size for the fixed-window case and Equation (9) for windows not exceeding age T .…”

CloudTCast: A Video Distribution Solution Based on Cloud and Tree-based ALM

“…As mentioned above, the tree-based ALM can obtain excellent forwarding effectiveness. In the mesh-pull P2P streaming protocols (e.g., [9], [10], [11]), the data is presented as a series of segments, each representing a short duration of playback. Every peer maintains a list of neighboring peers, and periodically exchanges segment availability information of data buffers (often referred to as buffer maps) with its neighboring peers.…”