Virtual Optical Network Mapping Approaches in Space-Division-Multiplexing Elastic Optical Data Center Networks

This paper addresses the problems of mapping virtual optical networks (VONs) onto space-division-multiplexing elastic optical data center networks (SDM-EODCNs). We first propose the concepts of link importance degree (LID) and node importance degree (NID) and put forward the calculation formula to optimize the VONs mapping, by which VON mapping approaches are developed. For comparison, we introduce a random (RAN) VON mapping approach that can provide benchmark performance. Simulation results show that the proposed LID VON mapping approach can yield solutions closest to the optimal ones obtained via an ILP model. On the other hand, compared to the NID and RAN VON mapping approaches, the LID VON mapping approach not only effectively improves the network profit by reducing the total network cost, but also solidly increases the number of VONs mapping.