Research On The Cache Management Strategy Of Streaming Media Based On P2P

The demand of streaming media quality of service is increasing along with the rapid development of network technology and the constant upgrade of personal computer configuration. Streaming media based on peer to peer network should be one good solution for this demand. The research on how to enhance the streaming media based on P2P system performance and quality of service has become the focus of current research. So it has the practical study meaning to research one efficient peer cache algorithm based on P2P streaming video on demand system.In this paper, we did a deep research on the control strategy of buffer cache based on peer to peer for video on demand system. The main work is as following:Firstly, this paper introduce the theory of caching technology and the importance that it apply in the VOD system based on P2P, which is based on theory introducing of P2P and streaming media. This paper also discussed the main factors that influence the quality of service of streaming media and the user visit character and behavior feature. In addition, it introduced the caching strategy and structure.Secondly, this paper presents one catch control strategy called SDCO(The combination of static and dynamic state) upon the compare of typical caching strategies. It divided the client buffer cache into three parts, they are fixed distribute cache, current play cache and prefetch distribute cache. All the peers have to have the fixed distribute cache to save the fixed contents after the peers join network successfully, and the peers can save or replace data in the benefit of play function or VCR operation according to the corresponding cache algorithm in other cache parts. SDCO have three parts, that is static data update algorithm, play data replacement algorithm and comprehensive evaluate determiner algorithm(CED), it considered the proportionality of the data blocks distribution and the hit rate of peer VCR operation comprehensively to enhance the quality of service.Finally, this paper illustrates the network simulation software NS2 in detail,and then simulate the SDCO strategy on it. The result indicates that the SDCO strategy is more efficient than the traditional strategy LRU in the aspects of server load, start delay and so on.