Video content delivery enhancement in CDNs based on users' social information

International audienceWith the huge growth of multimedia communication and digital content availability, energy efficient content delivery became an important research topic with the goal of reducing energy consumption of the intermediary nodes while providing better services and QoE to the end users. In this paper we focus on the subject of reducing the overall network energy consumption in accessing user generated content over social media platforms. We propose an approach, namely SocialiVideo which enables users to directly share their generated video content among existing social connections. We combine the approaches used in CDNs and P2P networks together with social connections between people in order to shorten the path the data traverses on average, and improve the latency. SocialiVideo places video content in users' premises (e.g., set-top-boxes) and serve others using a P2P connection. To this end, we use users' geolocation information retrieved from their network data (IP address) as well as subscribed social networks (e.g. current city attribute in Facebook) and social characteristics (e.g. friends list, and activities, etc.). In-order to evaluate the performance of the proposed solution, we implement our prototype based on Facebook/Akamai content delivery approach and evaluates the performance with reference to the current solution of the Facebook. Based on the results, SocialiVideo unload the traffic of the network and CDN, thus reduces network energy consumption and provides advantages for multiple entities, including CDNs and ISPs, as well as better QoE for end user