Resource availability–aware adaptive provisioning of virtual data center networks

摘要: Optimal provisioning of virtual networks with varying topologies on their physical counterpart is a major challenge in cloud data centers. Embedding the network shared infrastructure poses multitude challenges to provider. This paper proposes hybrid approach which adapts availability and fragmentation level residual resources for near‐optimal mapping multiple center substrate network. The proposed technique uses coordinated static dynamic embedding phases, adaptively, maximize density increases revenue service Also, it judiciously modifies existing mappings efficient utilization resources. Empirical results show that can attain ∼10% higher acceptance ratio compared exact heuristic techniques literature. With mixed integer programming model embedding, average path length reduced by 20% techniques.