Streaming Media Caching Model Based on Knapsack Problem

The dominant traffic on the Internet has changed from text and graphics based Web content to more information-rich streaming media content, such as audio and video. With the dramatic increase of network bandwidth and the advancement of technologies on media authoring, encoding, and distribution, media traffic on the Internet has increased explosively and now accounts for the majority of traffic volume. Modern Internet streaming services have utilized various techniques to improve the quality of streaming media delivery. Proxy server is one of the main solutions used to improve Internet QoS, especially for the QoS of streaming media. Replacement algorithm optimization is the core of caching model research. However, existing techniques for caching text and image resources are not appropriate for the rapidly growing number of continuous media streams. Based on the concept of hit ratio, this paper makes use of 0-1 knapsack problem to set up a hit ratio model of proxy cache, by use of which a proxy cache policy is presented. As compared with the classical dynamic streaming scheduling strategies, the proposed algorithm is shown that it can make full use of space of proxy cache, and also get a higher hit ratio.