Data Request And The Management Of Playback Buffer In P2P Streaming Media

Streaming media technical means that after compression video information is stored on the web server and users can download it, what's more, they need not wait until the entire compressed file is totally downloaded that only can watch it.Streaming media technology is not a single technology but a organic combination of network technical and audio techniques. In p2p streaming media system, the key issues include resource position,data request and the management of playback buffer. Because streaming media technology is based on the technology of p2p, so it also has inherent problems of p2p, such as safety problem,bandwidth consumption problem and identify problem, etc.The assessment of streaming media have several aspects, such as the rate of data request,started off and play delay, all of these aspects have direct connect with data request algorithm and buffer management algorithm both of which are the focus of this thesis.This thesis mainly focused on the data request and playback buffer management, discussed the time that the data blocks that have been played should be covered. Based on this, this thesis raised the idea of dynamic supply:demand, that is, when we assign coverage priority to data blocks that have been played, not only should consider the numbers of nodes that have got this data block and that haven't got but still need it, but also consider the locations of this block in the buffer of these nodes. In data management module, this thesis proposed newly joined node's starting request block should near one third of the greatest and most small number nodes among all blocks in it's all neighbor nodes' playback buffers.This thesis has made lots of contrast experiments respectively focusing on data request and management of playback buffer. The experiments' assessment standard is the data request rate, and they indicated that when newly joined node's starting request block near one third of the greatest and most small number nodes in all of it's neighbor nodes' playback buffers, the data request rate had been improved apparently. Besides this, when we assign coverage priority to data blocks that have been played with the idea of dynamic supply:demand, it also improved the data request rate.