A proxy caching algorithm based on popularity for streaming media

Abstract

A novel proxy caching algorithm P2CASM (Proxy Caching Algorithm Based on Popularity for Streaming Media) based on segment popularity for streaming media was proposed. A proxy caching admission and replacement algorithm based on the segment popularity for streaming media objects was implemented. The principle was obeyed that the data cached for each streaming media objects were in proportion to their popularity on the proxy server. The size of the caching window was updated periodically according to the average access time of the clients. P2P replica could be cached on a P2P proxy. Simulation results showed that this algorithm was more adaptive than A2LS (Adaptive and Lazy Segmentation Algorithm) algorithm for the variety of the proxy server cache. It could gain more average number of cached streaming media objects and less delayed requested ratio while byte-hit ratio of P2CASM algorithm was close to or exceeded A2LS algorithm under the circumstance of the same proxy cache space.