Sliding Window Caching Policy in Proxy Server for an Efficient Multimedia Streaming Service

Abstract

In this paper, we present a new cache management method, named Sliding Window Caching Policy to efficiently store the videos in a proxy server. The proxy server periodically measures time popularity (TP) and requested video segments from users. Based on these measured data, the proxy server calculates the average time popularity and average request segment of a video utilizing Sliding Window. Then, we estimate the priority of replacement when the cache is running out of space and the maximum size of storage space to be allocated to a video. In addition, we propose a conditional active prefetching in order to provide a continuous streaming delivery to clients. The simulation results show that the proposed algorithm performs better than other algorithm such as LRU, LFU, TP, Pyramid, and Prefix methods in terms of packet hit rate and number of replacement in random access pattern and Zipf distribution access Pattern.