WCSE 2019 SPRING ISBN: 978-981-14-1455-8
DOI: 10.18178/wcse.2019.03.024

Available Bandwidth Based Application-aware Engineering in SDN

May Thu Zar Win, Yutaka Ishibashi, Khin Than Mya

Abstract— Software-defined networking (SDN) has many advantages to support traffic engineering (TE) rather than the traditional network because of its decoupling of control and data plane, global centralized control, and enabling innovation through the network programmability. This paper focus es on the application-aware TE in SDN based on the estimat ion of end-to-end available bandwidth (ABW). First, we propose an application-aware TE framework for SDN which combines with sFlow-RT analyzer. The proposed framework performs three main tasks: traffic clas sification, t raffic measurement, and traffic management. As the prerequisite of port number based traffic classification, we use sFlow-RT which provides traffic met rics via REST API. Collecting traffic statistics and calculation of ABW for each link based on the provided met rics are managed by traffic measurement module. Traffic management performs flow rerouting based on calculated ABW and the application’s nature. We validate our proposed framework using the popular network emulat ion environment, Mininet and the widely used NOS called ONOS. Results of the experiment show increased throughput and decreased packet loss compared with other forwarding application.

Index Terms— Software-defined networking, traffic engineering, application-aware engineering, available bandwidth.

May Thu Zar Win, Khin Than Mya
University of Computer Studies, Yangon, MYANMAR
Yutaka Ishibashi
Nagoya Institute of Technology, JAPAN

[Download]


Cite: May Thu Zar Win, Yutaka Ishibashi, Khin Than Mya, "Available Bandwidth Based Application-aware Engineering in SDN," Proceedings of 2019 the 9th International Workshop on Computer Science and Engineering WCSE_2019_SPRING, pp. 142-147, Yangon, Myanmar, February 27-March 1, 2019.