PROJECT TITLE:

Video Delivery Performance of a Large-Scale VoD System and the Implications on Content Delivery

ABSTRACT:

Video delivery performance is the main issue that affects Net video quality. Characterizing the video delivery performance, especially the delivery throughput, will help content providers and Net service suppliers (ISPs) in system optimization and network designing. Based on a distinctive dataset consisting of twenty million video download speed measurements , this paper comprehensively studies the video delivery throughput of a large-scale industrial video-on- demand (VoD) system. We have a tendency to observe that user speed exhibits a massive variation over time of day as well as across provincial locations. In particular, the worst performance of day is thirty% lower than the peak performance . The analysis additionally reveals that video download speed contains a notable impact on Net video quality, which in flip influences user engagement . The impact, however, becomes restricted when the speed increases beyond a sure threshold, that is largely passionate about the video encoded bitrates. We tend to more examine the interaction between Net infrastructure and video delivery throughput using the linear regression model and find that crossing the ISP or regional network border yields fifteen–20p.c speed loss. Primarily based on these observations , we have a tendency to finally evaluate the potential of edge caching and hybrid CDN-P2P in the advance of video download performance and video quality.