As video streaming applications are deployed on the cloud, cloud providers are charged by ISPs for inter-data center transfers under the dominant percentile-based charging models. In order to minimize the payment costs, existing works expect to keep the traffic on each link under the charging volume. Be that as it may, these strategies can’t completely use each link’s available bandwidth capacity limit. As a solution, we propose an economical and deadline driven video flow scheduling system, called Eco Flow. Considering that different video streams have distinctive transmission dead lines, Eco Flow transmits videos in the request of their deadlines tightness and postpones the delivers of later-deadline video to later time slots.
The streams that are required to miss their deadlines are separated into sub flows to be rerouted to other under-used links. We likewise propose setting each link’s initial charging volume to lessen the booking dormancy toward the beginning of the charging time period and discuss how to manage with issues, for example, the forecast blunders of connection accessible transmission capacity and the absence of charging volume’s prior knowledge. Besides, we composed usage methodologies for utilizing Eco Flow in both centralized and distributed situations. Experimental results about show that Eco Flow accomplishes low bandwidth costs and higher video flow transmission rates when compared with existing strategies