Abstract
User Generated Content (UGC) videos have become very popular since the birth of web services such as YouTube and Youku allowing users to upload, view and share videos in an efficient manner. It is intriguing to study the distribution service of UGC videos for the rapidly increasing number of users with satisfactory Quality of Service (QoS). In this paper, we conduct extensive measurement of a real-world online UGC video system to study the properties of UGC videos. We summarize important characteristics in terms of correlation analysis, including temporal locality, geographic locality and propagation locality. We then propose a video popularity prediction framework based on a recurrent neural network. Furthermore, we design a content and network aware replication and scheduling mechanism, namely CNARS, to serve the distribution of UGC videos. Our trace-driven experiments demonstrate the effectiveness and superiority of CNARS, which significantly improves the hit ratio at edge data centers, lowers the average transmission distance and alleviates the server load for origin site against traditional approaches.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
