An Effective Data Scheduling Algorithm for Mesh-based P2P Live Streaming

Chunk scheduling is one of the main challenges and one of the keys to success for mesh-based Peer-to-Peer (P2P) streaming applications. Several chunk scheduling strategies have been proposed to achieve dissemination at an optimal rate or at an optimal delay. DLc strategy selects the minimum scheduling deadline to achieve optimal delay performance under neighbor-hood restrictive environments. By careful investigation of DLc, in this paper, we find that DLc algorithm doesn't deal with the condition when two or more chunks have the same scheduling deadline and only randomly selects one to send, which will descend its delivery efficiency. We then propose a hybrid-DLc scheduler aiming to provide the solution when chunk scheduling meets this condition. And we show, through a comprehensive set of simulation with realistic scenarios, that this scheduler outperforms DLc in any scenario experimented.