On Advantages of Traffic Prediction and Grooming for Provisioning of Time-Varying Traffic in Multilayer Networks

The constantly increasing traffic volume triggers the fast development of new optimization methods for backbone optical networks. Operators seek new ways to improve several network performance metrics, including network stability. In this paper, we propose a multilayer packet-over-optical approach combining grooming and traffic prediction. We demonstrate the advantages of the proposed algorithm for provisioning daily time-varying traffic of various services through multiple metrics. We show how the proposed approach, enabled by the cross-layer information exchange, can achieve a served traffic increase of about 26% in operational scenarios.