Optical network automation [Invited tutorial]

© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.Increased levels of automation will be necessary in view of more stringent performance requirements that next generation optical transport networks need to support in a near term, not only for high capacity, but, even more importantly, for dynamicity, latency, and availability. All these aspects will become more relevant with the growing complexity of modern networks. Network automation targets resource reoptimization to rapidly adapt the network to the expected conditions, quick degradation detection to improve the quality of the connections, as well as failure detection and identification to facilitate maintenance. Network automation requires and implies the collection of data for performance monitoring, being then elaborated by data analytics algorithms to produce meaningful inputs for the network controller, which will finally program the underlying devices. In this paper, we analyze alternative architectures for monitoring and data analytics (MDA) and illustrative control loops are presented aiming at validating the usefulness of MDA to automate optical networks operation.The research leading to these results has received funding from the Spanish MINECO TWINS project (TEC2017-90097-R), and from the Catalan Institution for Research and Advanced Studies (ICREA).Peer ReviewedPostprint (author's final draft