Taxonomy of traffic engineering mechanisms in software-defined networks: a survey

Mohammadi R., Akleylek S., Ghaffari A., Shirmarz A.

TELECOMMUNICATION SYSTEMS, vol.81, no.3, pp.475-502, 2022 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 81 Issue: 3
  • Publication Date: 2022
  • Doi Number: 10.1007/s11235-022-00947-6
  • Journal Name: TELECOMMUNICATION SYSTEMS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, ABI/INFORM, Applied Science & Technology Source, Business Source Elite, Business Source Premier, Communication Abstracts, Compendex, Computer & Applied Sciences, INSPEC
  • Page Numbers: pp.475-502
  • Keywords: SDN, Traffic engineering, Traffic measurement, Traffic management, Load balancing, QoS, TOPOLOGY DISCOVERY, LOW-LATENCY, SDN, QOS, MANAGEMENT, PROTOCOL
  • Ondokuz Mayıs University Affiliated: Yes

Abstract

Nowadays, many applications need varying levels of Quality of Service (QoS). The network that provides the communication service connects the servers and clients. The network traffic which is routed through the network should be engineered. Traffic Engineering (TE) is a mechanism for transferring the packets considering the different QoS level requirements among applications. The optimal resource allocation is the primary strategy for TE so that the network can provide the QoS requirements for each application. The TE can improve network efficiency, performance, and user satisfaction. Software Defined Network (SDN) has been proposed as the novel network architecture that could make networks agile, manageable, and programmable using control and data plane separating compared to traditional network architecture. In this paper, we survey network traffic engineering in SDN. We investigate and cluster the articles published between 2017 and 2022 on traffic engineering in SDN. The state-of-the-art articles about the traffic engineering mechanisms in SDN have been examined and classified into four types: topology discovery, traffic measurement, traffic load balancing, QoS, and dependability. Finally, the cutting-edge issues and challenges are discussed for future research in SDN-based TE.