Improving Streaming Capacity in Multi-ChannelP2P VoD Systems via Intra-Channel andCross-Channel Resource Allocation

摘要

Multi-channel Peer-to-Peer (P2P) Video-on-Demand(VoD) systems can be categorized intoindependent-channelP2PVoD systems andcorrelated-channelP2P VoD systems. Streamingcapacity for a channel is defined as the maximal streamingrate that can be received by every user of the channel. In thispaper, we study the streaming capacity problem in multi-channelP2P VoD systems. In an independent-channel P2P VoD system,there is no resource correlation among channels. Therefore, wecan find the average streaming capacity for the independent-channelP2P VoD system by finding the streaming capacity foreach individual channel, respectively. We propose a distributedalgorithm to solve the streaming capacity problem for a singlechannel in an independent-channel P2P VoD system. The averagestreaming capacity for a correlated-channel P2P VoD systemdepends on both the intra-channel and cross-channel resourceallocation. To better utilize the cross-channel resources, wefirst optimize the server upload allocation among channels tomaximize the average streaming capacity and then proposecross-channel helpers to enable cross-channel sharing of peerupload bandwidths. We demonstrate in the simulations that thecorrelated-channel P2P VoD systems with both intra-channel andcross-channel resource allocation can obtain a higher averagestreaming capacity compared to the independent-channel P2PVoD systems with only intra-channel resource allocation.