An efficient embedding algorithm for virtual network via exploiting topology attributes and global network resources

Abstract

Virtual Network Embedding (VNE) is one of the longstanding challenges in Network Virtualization: how to efficiently embed multiple virtual networks (VNs), with node-link resource requirements, onto the shared substrate network, having finite underlying resources. Proposed algorithms are heuristic in the literature. Most heuristic VNE algorithms, only considering single network topology attribute and local node resources to rank nodes, will lead to low resource utilization of substrate network and low average VN acceptance ratio in the long term. To deal with this issue, this paper proposes another efficient heuristic algorithm VNE-TAGNR, adopting a novel node-ranking approach in the node mapping stage and coordinating the following link mapping stage. In the node-ranking approach, three fundamental Topology Attributes and Global Network Resources (TAGNR) are considered and quantified. Numerical simulation results vividly reveal that the proposed VNE-TAGNR algorithm outperforms four typical and state-of-the-art heuristic algorithms, in terms of average VN acceptance ratio, average node / link utilization and average revenue / cost.