Location-aware energy efficient virtual network embedding in software-defined optical data center networks

• Published in: Journal of optical communications and networking Vol. 10; no. 7; pp. 58 - 70
• Main Authors: Zong, Yue, Ou, Yanni, Hammad, Ali, Kondepu, Koteswararao, Nejabati, Reza, Simeonidou, Dimitra, Liu, Yejun, Guo, Lei
• Format: Journal Article
• Language: English
• Published: Piscataway Optica Publishing Group 01.07.2018; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Ant colony optimization; Computer centers; Computer simulation; Data center networks; Data centers; Embedding; Energy costs; Energy efficiency; Integer programming; Linear programming; Mixed integer; Optical communication; Optical fiber networks; Optical fibers; Power consumption; Power demand; Resource allocation; Servers; Software; Software defined optical; Virtual network embedding; Virtual networks
• ISSN: 1943-0620; 1943-0639
• DOI: 10.1364/JOCN.10.000B58

To overcome the Internet ossification, network virtualization has been proposed as a promising method because of its advantages (e.g., on-demand and efficient resource allocation). Virtual network embedding (VNE) is one of the main challenges for network virtualization. Energy costs of servers in data centers (DCs) aremajor contributions to the power consumption in information and communication technology. Therefore, VNE should consider both acceptance ratio and power consumption. In this paper, a mixed integer linear programming model is proposed with the objective of minimizing the total power consumption in software- defined optical data center networks by reducing the active data centers and power-consuming network components. In addition, the coordinates of nodes and delay of links are considered for a more realistic scenario. Compared with the existing node ranking method, the proposed global topology resource (GTR) can effectively evaluate the possibility of each DC node to host virtual nodes. Based on the GTR method, we propose a location-aware energy efficient VNE algorithm, namely GTR-VNE. Simulation results show that GTR-VNE can obtain up to 9.3% and 5% improvement of power consumption and acceptance ratio compared with benchmarks. Furthermore, based on GTR and artificial intelligence ant colony optimization (ACO), another energy efficient algorithm, ACO-VNE, is proposed.ACO-VNE can obtain up to 28.7% improvement in power consumption compared with GTR-VNE. In addition, ACO-VNE has better performance in terms of revenue cost ratio and acceptance ratio.