Joint Node-Link Embedding Algorithm based on Genetic Algorithm in Virtualization Environment

Virtual network embedding (VNE), that efficiently tackles the mapping problems of heterogeneous virtual networks onto a shared physical infrastructure meeting rigid resource constraints, is the major challenge in network virtualization (NV). VNE is widely known as NP-hard due to its intractable computation. The majority of VNE solutions have concentrated upon virtual node mapping (VNoM) and virtual link mapping (VLiM) separately. Uncoordinated approaches would facilitate algorithmic implementation, but they lead to low acceptance ratio, network revenues and high embedding cost. In this paper, we propose a new approach relied on Genetic Algorithm (GA), that coordinately joints node and link mappings where the link embedding is based on a fast and efficient sequential path searching method. A novel heuristic conciliation algorithm is presented to deal with a set of infeasible link mappings during producing VNE solutions in GA's operations. Extensive evaluation results indicate that our proposed approach outperforms state-of-the-art VNE algorithms in all adopted performance metrics.