Research And Implementation Of Router Acceleration Based Strategy For Streaming Delivery

With the rapid development of Internet, many applications based on streaming media technology, such as VOD, distance education, video conferencing, video telephony, have become more and more popular in the Internet. For this reason, streaming media distribution technologies become increasingly important. Due to data volume, low latency and low jitter characteristics of streaming media, streaming media applications have a high demand on network bandwidth and server performance, while the existing Internet cannot meet the high requirements of streaming media applications.It is an effective way of pushing media content to the edge of the network or sharing resources between clients to address the problem that the high requirements of streaming media applications does not match with existing network performance. CDN (Content Delivery Network) and P2P (Peer-to-Peer) are two relatively mature technologies. In content delivery network, a large number of CDN servers are deployed at the edge of the network to enhance the service capabilities of the system. However, it is expensive to deploy such a large number of servers to limit the system scalability. P2P technology fully use client resources to reduce the load of the streaming media server, but it brings an huge burden on the ISP.Another idea to solve the problem is to build a new network architecture, which focuses on content rather than on the transmission. CCN (Content Centric Networking) is a new network architecture like that, in which content can be cached throughout the network, as long as it is required, the content can be effectively distributed at anytime, anywhere. Router, as the key equipment in the network, of course should take the task of caching and distribution.In this paper, we analyze the difference between edge routers and core routers and we think the edge routers have more spare computing and storage resources to assume the task of caching and distribution of streaming media, and then we have proposed a strategy of media content delivery acceleration based on the edge routers. In the strategy, edge routers are organized into different groups and take the initiative to cache popular video data. Then when the user requests media content which in the cache, the edge routers can directiy delivery media content to the user. Then we have designed some algorithms to improve the utilization of the cache on the edge routers. At last, we design a prototype system of VOD based on the acceleration strategy. With the experiment results, it is evidenced that the edge router can significantly accelerate streaming media distribution, and effectively decrease the streaming media applications’demand on the network bandwidth requirements and server capacity. We think the media delivery acceleration method based on the edge routers is applicable to large-scale streaming media content distribution, and it is significant for building the next generation network.