Energy-efficient virtual network embedding in networks for cloud computing. (2017)

• Record Type: Journal Article
• Authors: Zheng, Xiang-Wei; Hu, Bin; Lu, Dian-Jie; Chen, Zhen-Hua; Liu, Hong
• Is Part Of: International journal of web and grid services. Volume 13:Number 1(2017)
• Page Start: 75
• Page End: 93
• Publication Date: 2017
• Subjects: cloud data centres -- network as a service -- NaaS resource allocation -- virtual networks -- embedded systems -- energy consumption -- group search optimisation -- energy efficiency -- cloud computing -- cloud services -- virtual resources -- network virtualisation -- simulation; Web services -- Periodicals; Computational grids (Computer systems) -- Periodicals; 006.78
• Languages: English
• ISSNs: 1741-1106

Abstract

Cloud computing is based on several service models such as Network as a Service (NaaS), Software as a Service (SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS). NaaS is a business model that aims to provide virtual network (VN) services over the internet from shared cloud data centres. Virtual network embedding (VNE) is one of the core technologies in NaaS resource allocation. In view of the enormous energy consumption by numerous cloud data centres, energy-efficient VNE becomes a new research focus. In this paper, we propose an energy efficient virtual network embedding (EEVNE) approach for cloud computing networks, in which power savings are introduced by consolidating resources in the network and data centres. EEVNE is based on the evaluation of the energy consumption by hosts, the allocation of virtual resources both in nodes and links, as well as the cost and revenue in network virtualisation. We also formulated a heuristic embedding algorithm based on Group Search Optimiser (GSOVNE) to solve the NP-hard problem. Furthermore, a reconfiguration algorithm is developed to improve the utility of fragmented resources. Simulation experiments show that the proposed algorithm can effectively increase VN acceptance ratio and reduce energy consumption when large quantities of virtual networks arrive and depart over time.