Virtual Network Embedding Using Node Multiple Metrics Based on Simplified ELECTRE Method
Open Access
- 15 June 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 6, 37314-37327
Abstract
The concept of network virtualization has attracted significant attention from academia to industry. One of the key challenges in network virtualization is the resource allocation problem, which is also termed the virtual network embedding (VNE) problem. It involved with mapping virtual networks onto a substrate network by adhering to some constraints, such as CPU capacity, on the nodes and bandwidth resources on the links. However, prior heuristic VNE algorithms mostly concentrate on measuring the embedding potential of substrate nodes using the multiplication of different nodes' resource metrics. Due to the fact that different resource metrics have different impacts on node ranking, these traditional methods have some limitations that would cause unbalanced embedding problems. Furthermore, the number of hops for the substrate paths that virtual links are mapped onto will have a large impact on the resource utilization of substrate links in a substrate network. In this paper, based on the topology analysis of six situations, we first propose a novel five-node ranking metric to quantify the importance of substrate nodes. Then, we give a comprehensive measurement method for substrate nodes using the simplified ELECTRE method to avoid an unbalanced embedding solution. We present a novel two-stage VNE algorithm, which chooses the substrate nodes with the maximum embedding potential to perform the node mapping procedure, and uses the shortest path algorithm to accomplish the link mapping procedure. Extensive simulation results demonstrated that our proposed method behaves better than the other state-of-the-art algorithms in terms of the long-term average revenue, the revenue-to-cost (R/C) ratio, and the VN request acceptance ratio.Keywords
Funding Information
- Fundamental Research Funds for the Central Universities of the China University of Petroleum (East China) (18CX02139A)
- Natural Science Foundation of Shandong Province (ZR2014FQ018)
- National Natural Science Foundation of China (61471056)
- National Basic Research Program (973) of China (2012CB315801)
- China Research Project on Key Technology Strategy of Infrastructure Security for Information Network Development
This publication has 27 references indexed in Scilit:
- Virtual network embedding based on real-time topological attributesFrontiers of Information Technology & Electronic Engineering, 2015
- A memetic algorithm for the virtual network mapping problemJournal of Heuristics, 2014
- Greener networking in a network virtualization environmentComputer Networks, 2013
- A unified enhanced particle swarm optimization‐based virtual network embedding algorithmInternational Journal of Communication Systems, 2012
- Virtual network embedding through topology-aware node rankingACM SIGCOMM Computer Communication Review, 2011
- Virtual network provisioning across multiple substrate networksComputer Networks, 2011
- Rethinking virtual network embeddingACM SIGCOMM Computer Communication Review, 2008
- An overview on “The European school of MCDA: Emergence, basic features and current works”European Journal of Operational Research, 1997
- Optimization by Simulated AnnealingScience, 1983
- Multi-criteria Analysis and Regional Decision-Making.Journal of the American Statistical Association, 1980