PROJECT TITLE :

Coping With Heterogeneous Video Contributors and Viewers in Crowdsourced Live Streaming: A Cloud-Based Approach

ABSTRACT:

With the advances in personal computing devices and the prevalence of broadband network and wireless mobile network accesses, finish-users are now not pure content consumers, but contributors, too. In today's crowdsourced streaming systems, varied broadcasters spirited stream their video content, e.g., live events or online game scenes, to fellow viewers. Compared to professional video producers and broadcasters, these new generation broadcasters are geo-distributed globally and highly heterogeneous in terms of the generated video quality and the network/system configurations. The scalability and heterogeneity challenges so lie on each broadcasters and therefore the viewers, which call for huge transcoding, and 2 vital issues: one) selecting video illustration set that maximizes viewer satisfaction and a couple of) allocating computational resources that minimize operational prices, must be systematically optimized in the global scale. In this paper, we present a generic framework utilizing the powerful and elastic cloud computing services for crowdsourced live streaming with heterogeneous broadcasters and viewers. We have a tendency to jointly take into account the viewer satisfaction and also the service availability/pricing of geo-distributed cloud resources for transcoding. We tend to develop an optimal scheduler for allocating cloud instances with no regional constraints. We have a tendency to then extend the answer to accommodate regional constraints, and discuss a series of sensible enhancements, as well as popularity forecasting, initialization latency, and viewer feedbacks. Our solutions have been evaluated under numerous networks and cloud system configurations with parameter settings. The trace-driven simulation confirms the prevalence of our design, whereas our Planetlab-primarily based experiment offers more practical hints toward real-world migration.