An Energy Efficient Embedding Algorithm for Multi-Domain Virtual Network Embedding

Abstract

Network virtualization (NV) is one fundamental attribute of the next generation network. Virtual network embedding (VNE) is about the resource allocation in NV topic. Researchers have proposed multiple embedding algorithms since 2008. Preceding algorithms try to accept more virtual networks (VNs) in one underlying substrate network (SN) and maximize embedding revenue. As energy cost accounts for approximately half whole network cost, it is essential for Internet service providers to optimize energy cost. In addition, in real networking environment, it is vital to embed VNs among multiple SNs, geographically distributed in different regions. To solve both issues, we firstly propose the Multi-Domain VNE problem model and formal energy cost model for VNE. Then, we present a two-stage and energy efficient algorithm (EEMD-VNE-T), aiming at optimizing energy cost. Meanwhile, the EEMD-VNE-T is able to embed each VN among multiple SNs in polynomial time. Preliminary results reveal that EEMD-VNE-T significantly reduces the energy cost by roughly 8.7% over two representative energy algorithms.