University of Surrey

Test tubes in the lab Research in the ATI Dance Research

On IGP link weight optimization for joint energy efficiency and load balancing improvement

Francois, F, Wang, N, Moessner, K, Georgoulas, S and Xu, K (2014) On IGP link weight optimization for joint energy efficiency and load balancing improvement Computer Communications, 50. pp. 130-141.

[img] Text
Restricted to Repository staff only
Available under License : See the attached licence file.

Download (6MB)
[img] Text (licence)
Available under License : See the attached licence file.

Download (33kB)
on IGP link weight.pdf - ["content_typename_UNSPECIFIED" not defined]
Available under License : See the attached licence file.

Download (524kB) | Preview


The energy consumption of backbone networks has risen exponentially during the past decade with the advent of various bandwidth-hungry applications. To address this serious issue, network operators are keen to identify new energy-saving techniques to green their networks. Up to this point, the optimization of IGP link weights has only been used for load-balancing operations in IP-based networks. In this paper, we introduce a novel link weight setting algorithm, the Green Load-balancing Algorithm (GLA), which is able to jointly optimize both energy efficiency and load-balancing in backbone networks without any modification to the underlying network protocols. The distinct advantage of GLA is that it can be directly applied on top of existing link-sleeping based Energy-aware Traffic Engineering (ETE) schemes in order to achieve substantially improved energy saving gains, while at the same time maintain traditional traffic engineering objectives. In order to evaluate the performance of GLA without losing generality, we applied the scheme to a number of recently proposed but diverse ETE schemes based on link sleeping operations. Evaluation results based on the European academic network topology GÉANT and its real traffic matrices show that GLA is able to achieve significantly improved energy efficiency compared to the original standalone algorithms, while also achieving near-optimal load-balancing performance. In addition, we further consider end-to-end traffic delay requirements since the optimization of link weights for load-balancing and energy savings may introduce substantially increased traffic delay after link sleeping. In order to solve this issue, we modified the existing ETE schemes to improve their end-to-end traffic delay performance. The evaluation of the modified ETE schemes together with GLA shows that it is still possible to save a significant amount of energy while achieving substantial load-balancing within a given traffic delay constraint. © 2014 Elsevier B.V. All rights reserved.

Item Type: Article
Divisions : Faculty of Engineering and Physical Sciences > Electronic Engineering > Centre for Communication Systems Research
Authors :
Date : 1 September 2014
Identification Number : 10.1016/j.comcom.2014.05.003
Additional Information : NOTICE: this is the author’s version of a work that was accepted for publication in Computer Communications. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Computer Communications, 50, September 2014, DOI 10.1016/j.comcom.2014.05.003.
Depositing User : Symplectic Elements
Date Deposited : 13 Feb 2015 10:30
Last Modified : 13 Feb 2015 14:33

Actions (login required)

View Item View Item


Downloads per month over past year

Information about this web site

© The University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom.
+44 (0)1483 300800