PROJECT TITLE :

Collaborative Repositories in Model-Driven Engineering [Software Technology]

ABSTRACT:

Recent years have witnessed video-on-demand (VoD) as an economical means for providing reliable streaming service for Web users. It's known that peer-assisted VoD systems, like NetFlix and PPlive, typically incur a lower deployment cost in terms of server bandwidth consumption. However, some basic problems still would like to be more clarified, particularly for VoD service suppliers. In particular, how so much will we push peer-assisted VoD forward, and at the scale of VoD systems, the most reduction of server bandwidth consumption which will be achieved with peer-assisted approaches. During this paper, we have a tendency to provide in depth model analysis to perceive the minimum server bandwidth consumption for peer-assisted VoD systems. We tend to first propose a basic model that may optimally schedule user demands at given snapshots. Our model analysis reveals the optimal performance sure and shows that the prevailing peer-assisted protocols are still way from being optimal. How to push the server bandwidth consumption to the limit remains a massive challenge in VoD system design. To approach the optimal bandwidth consumption in real deployment, we have a tendency to further extend our model to a practical case to capture the peer dynamic across continuous time-slots. The simulation result indicates that the optimal load scheduling downside continues to be achievable through a dynamic programming algorithm. Its design principle further motivates a quick priority-based mostly algorithm that achieves near-optimal performance. These proposed algorithms will considerably reduce the bandwidth consumption of dedicated VoD servers.