Facial reconstruction

Search LJMU Research Online

Browse Repository | Browse E-Theses

An Adaptive and Lightweight Update Mechanism for SDN

Yu, C, LAN, J, Guo, Z, Hu, Y and Baker, T (2019) An Adaptive and Lightweight Update Mechanism for SDN. IEEE Access, 7. pp. 12914-12927. ISSN 2169-3536

Full text not available from this repository. Please see publisher or open access link below:
Open Access URL: http://dx.doi.org/10.1109/ACCESS.2019.2893058 (Published version)

Abstract

To improve the network resources utilization and the quality of service, the provision of an adaptive and customizable network service is deemed a feasible approach. In this paper, based on the Quality of Service (QoS)-aware traffic classification and real-time network status, an adaptive update mechanism is presented to change the traditional rigid update techniques in SDN. The developed update mechanism aims at abstracting the common update mechanism into update operations and calculates the update sequence on the operation granularity. The mechanism has three work modes, and each mode has a corresponding algorithm. It can adjust the work modes adaptively based on the network condition and the flow QoS requirements to improve the performance. The experimental results demonstrate that the three work modes can achieve optimal performance in Ternary Content Addressable Memory (TCAM) overhead reduction, delay, and bandwidth consumption, respectively. For example, when the Tri-fusion work mode is leveraged, it provides at least an 85% reduction of the additional TCAM overhead and improves by at least 9%, 65%, and 82% compared to other work modes and the compared algorithms.

Item Type: Article
Additional Information: © 2019 IEEE
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Computer Science
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date Deposited: 15 Jan 2019 10:40
Last Modified: 24 May 2019 10:34
DOI or Identification number: 10.1109/ACCESS.2019.2893058
URI: http://researchonline.ljmu.ac.uk/id/eprint/9946

Actions (login required)

View Item View Item