The Research And Apply Of Resource Schedule Of A Streaming Media System

Streaming media technology has been widely applied in many areas of internet information service such as multimedia news release, live video playing, net ads, VOD, long-distance education, long-distance medical treatment and so on now. On the net, resources which can provide us with streaming media service are comparatively limited. Thus, how to offer enough streaming media service with a certain quality guarantee to the growing amount of streaming users has become a matter of great concern. At the same time, the resource schedule policy of streaming media system has been developed to solve the problem, Proper resource schedule algorithm could help the streaming media system to improve its service ability. Naturally it has become the key technology as well as the major research area in this system.This thesis is aimed at the research on the resource schedule of the streaming media system. First, the basic knowledge of streaming media technology and the significance of the resource schedule are specifically researched here. And then the resource schedule is logically divided into three aspects: the storage policy, the playing algorithm about server to client and the streaming schedule policy among servers. Current achievements have been analyzed and summarized. we have absorbed the merits of these aspects and pointed out their limitations.Based on these researches and a practical system of streaming media, a set of schemes to the resource schedule has been proposed in this thesis. By using program segment, distributed storage and broadband media distribution protocol, the schemes are able to store and swift over the program automatically on the net. unicast, multicast and the exchange between unicast and multicast; and near VOD algorithm has been adopted to play the streaming. And in order to keep the load balance between streaming media servers when parallel client number request, a new policy to schedule the streaming has been created here which adopts the method of "Instructions-intensive solution" and introduces a new concept called "tentative count". Based on the combination of "tentative count" and "practical count", the streaming media system will get the idea on how to distribute the streaming servers. The server with program segment on request and the smallest amount of streaming will be supplied with streaming service.Then, the new policy has been tested. Based on the test theory, the test parameters, test tools and test method about new streaming schedule policy have been researched and identified. The test result shows that the new policy can support much more clients and improve the system performance effectively.At last, the works of this thesis have been summarized and the further research works on resource schedule of streaming media system have been pointed out.