Network-assisted HTTP adaptive streaming for hybrid software defined network

Abstract

With the ever increasing of Internet video traffic, HTTP adaptive streaming (HAS) is emerging as the de facto standard for video streaming. Many works based on software defined networking (SDN) have been proposed to improve HAS end-user quality of experience (QoE). While HAS can benefit greatly from SDN centralized controller - maintaining better network resource utilization and overall quality stability ensuring fairness between clients, it also has to deal with SDN's scalability problem which makes a full SDN deployment difficult. Hybrid-SDN, which combines SDN and traditional network is a potential solution to this problem. However, to the best of our knowledge, none of the existing works can be fully operated in a Hybrid-SDN environment. In this paper, we propose a novel network-assisted HAS system for Hybrid-SDN that can provide comparable end-user QoE to conventional approach. Our architecture alleviates the problem of SDN controller having partial view of the network by estimating HAS clients network conditions based on information from the egress link connecting SDN and traditional network. Experimental results show the feasibility of our approach.