Research On Caching Technologies For Media Distribution Based On Content Popularity

With popularizing the typical stream system, such as Video-on-Demand, Video Conference and E-Learning etc, the multimedia traffic takes the largest portion in the current Internet traffic. So, the challenge is to provide multimedia services efficiently and scalable. To date, multimedia distribution services basically fall into three categories based on architecture: 1) centralized multimedia distribution; 2) content distribution network (CDN)-based multimedia distribution; and 3) peer-to-peer (P2P) networks. With varying characters and facing diverse challenge, they are dealt with differently.This thesis addresses on the cache scheme for media distribution under CDN-based and P2P network-based architecture. For CDN-based architecture, because the resource such as memory is limited, we should make use of them efficiently to improve the performance of systems. For P2P-based architecture, every node in the network will contribute part of its resource, and play as not only client but also server. How to aggregate and utilize the scattered resource efficiently is the key to this architecture. Memory is also the most important resource for such system, so the cache scheme will be exploited. The works in this thesis has been supported by the Natural Science Foundation of China"Internet-based interactive streaming delivery technique"(No.60302004) and Natural Science Foundation of Hubei"Stream media caching based on quality and time two-dimension attribute"(No.2005ABA264).The contributions of this thesis include:(1)With the limitation of cache space, improve the stream media transmission quality and lower the consumption of backbone bandwidth is a conflict. A new stream media caching and transmission scheme which combines with FGS coding has been proposed in this thesis.(2)Through introducing a novel conception Local Content Popularity, we extend the definition of Content Popularity from one-dimension of media object to two-dimension of media object and special area. Based on this conception, we modify and enhance a well-known cache scheme, then make a detail comparison between them.