Design of an Efficient Resource Management Model in Elastic Optical Networks

Abstract

The advent of elastic optical networks (EONs) has drawn much attention in recent years and this new paradigm is identified as the most promising solution to the next generation optical networks. In EONs, routing and spectrum assignment (RSA) problem plays a role very similar to the routing and wavelength assignment (RWA) problem in wavelength-routed networks. As the RSA problem has been proved to be NP-hard, how to design an efficient resource management model to improve execution efficiency of RSA heuristics would become a practical and critical issue. In this work, we address a whole new resource management unit named cluster to simplify the spectrum assignment step. Because of the trade-off between computation efficiency and blocking performance, it is expectable that the cluster-based RSA algorithm would suffer from the problem of low spectral utilization. To effectively alleviate the inevitable intra-and inter-cluster fragmentation, we also propose some additional processing options as possible solutions. Numerical results show that the proposed scheme is a highly competitive dynamic RSA algorithm because of its acceptable blocking performance and significantly improved execution efficiency as compared with near-optimal algorithm. It reveals that cluster-based resource management model can not only achieve the goal of good time efficiency but also maintain satisfactory blocking performance.