Title :
A new pre-pushing VoD scheme in hierarchical network environment
Author :
Fei Long ; Xingjun Wang
Author_Institution :
Dept. of EE, Tsinghua Univ., Shenzhen, China
Abstract :
VoD (Video-on-Demand) has become a hot topic over the last few years. Research was conducted on reducing the bandwidth cost of it. Different from conventional schemes like batching and patching, pushing popular videos to users with external storage during off-peak periods gives a new perspective. In this paper, we propose a new scheme under a hierarchical network environment, named “PPVoD (Pre-Pushing VoD)”. We segment programs into chunks and predict popularities of them in advance. According to the results, we work out a greedy algorithm to decide how to push the chunks to each level. And we develop a method using system cost to evaluate the efficiency of the PPVoD scheme. Simulation results suggest that the PPVoD scheme makes sense in real VoD systems.
Keywords :
greedy algorithms; image segmentation; video on demand; video signal processing; PPVoD; bandwidth cost reduction; external storage; greedy algorithm; hierarchical network environment; prepushing VoD scheme; program chunks; real VoD systems; segmentation; system cost; video popularities; video-on-demand; Analytical models; Bandwidth; Educational institutions; Greedy algorithms; Internet; Servers; Videos; VoD; hierarchical; popularity; pre-pushing; segmentation;
Conference_Titel :
Data Science and Advanced Analytics (DSAA), 2014 International Conference on
DOI :
10.1109/DSAA.2014.7058073