In cloud infrastructure, accommodating multiple virtual networks on a singlephysical network reduces power consumed by physical resources and minimizescost of operating cloud data centers. However, mapping multiple virtual networkresources to physical network components, called virtual network embedding(VNE), is known to be NP-hard. With considering energy efficiency, the problembecomes more complicated. In this paper, we model energy-aware virtual networkembedding, devise metrics for evaluating performance of energy aware virtualnetwork-embedding algorithms, and propose an energy aware virtualnetwork-embedding algorithm based on multi-objective particle swarmoptimization augmented with local search to speed up convergence of theproposed algorithm and improve solutions quality. Performance of the proposedalgorithm is evaluated and compared with existing algorithms using extensivesimulations, which show that the proposed algorithm improves virtual networkembedding by increasing revenue and decreasing energy consumption.
展开▼
