1570369620.pdf (805.48 kB)
Proactive admission control and dynamic resource management in SDN-based virtualized networks
conference contribution
posted on 2017-09-21, 15:42 authored by Sara Shakeri, Saeedeh Parsaeefard, Mahsa DerakhshaniNetwork virtualization is a promising approach in which common physical resources are shared between service
providers. Due to the substrate network limitations such as maximum
available memory of each node of the substrate network as well as different service priorities and requirements, resource management in this setup is essential. On the other hand, SDN
is bringing a considerable flexibility in resource management by introducing a centralized controller which can monitor all the substrate network states. In this paper, we propose a proactive admission control and dynamic resource management in SDNbased virtualized network in which the number of accepted highpriority virtual network (VN) requests is maximized, subject to both substrate limitations and memory requirement of each VN request. In the proposed formulation, based on the prediction of the substrate network utilization, we reserve resources for
upcoming high-priority VN requests. Via simulation, we show that the algorithm can increase the acceptance ratio of the highpriority VN requests up to % 100 where the substrate network is congested, i.e., arrival rates of both high-priority and low-priority VN requests are high.
History
School
- Mechanical, Electrical and Manufacturing Engineering
Published in
Network of the Future (NoF)Citation
SHAKERI, S., PARSAEEFARD, S. and DERAKHSHANI, M., 2018. Proactive admission control and dynamic resource management in SDN-based virtualized networks. IN: 8th international Conference on the Network of the Future (NOF), London, UK, 22-24 November 2017.Publisher
© IEEEVersion
- AM (Accepted Manuscript)
Acceptance date
2017-08-24Publication date
2018Notes
© 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.ISBN
9781538605547Publisher version
Language
- en