DASH-QoS: A scalable network layer service differentiation architecture for DASH over SDN

Sayit M., Cetinkaya C., Yıldız H. U., Tavli B.

COMPUTER NETWORKS, vol.154, pp.12-25, 2019 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 154
  • Publication Date: 2019
  • Doi Number: 10.1016/j.comnet.2019.02.015
  • Journal Name: COMPUTER NETWORKS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.12-25
  • Keywords: HTTP Streaming, Service classes, Openflow, MIP, Heuristics, CENTRIC NETWORKING, VIDEO, FAIRNESS
  • TED University Affiliated: Yes

Abstract

As one of the key technologies residing in the future Internet concepts, Software Defined Networking (SDN) makes it possible to develop application-centric flow routing strategies. Thanks to the benefits offered by the SDN architecture, the performance of multimedia communication applications can be improved by developing application-centric flow routing strategies. Recently, MPEG-DASH standard which allows rate adaptation over HTTP has become popular. In this work, we present an architecture that provides increase in received quality of DASH clients running over an SDN domain. Within this architecture, we propose an optimization model which determines the streaming paths of the video packets of the clients. We also define different types of user classes and provide a guaranteed service and fairness among the users of the same class. Furthermore, we created a heuristic algorithm to facilitate the scalability of the solution. The simulation results show that, by considering the output of the optimization model and dynamically changing the routing paths in an ISP, the video quality offered to the DASH clients is significantly improved and a certain level of service guarantee is provided. (C) 2019 Published by Elsevier B.V.