Optical metro networks evolution driven by 5G requirements face enormous challenges. Network functions virtualized in the data centers spread to the metro nodes, IP, and optical technologies must cooperate to meet the metro traffic aggregation role. Multiple technological options exist, and carriers confront the need to economically assess them, benchmarked in realistic deployments. This paper gives relevant insights to this aim. We first construct a set of metro network benchmarks. A strategic and distinctive effort is made to incorporate metro WDM topologies, traffic profiles and daily variation patterns, fault-tolerance requisites, and network operational choices, that faithfully reflect the expected 5G metro progression for a national carrier. Then, we use these networks to assess two technological choices. On one hand, the cost-effectiveness limits in terms of CAPEX reductions and energy efficiency brought from the possibility of having an agile control plane in the metro, capable of on-demand instantiation of IT and network resources. On the other hand, we investigate the benefits of replacing ROADMs by more cost-effective filterless technologies, but just limiting this replacement to degree-1 and degree-2 optical nodes, that are prevalent (e.g. >50%) in regional metro topologies. A novel capacity planning algorithm has been developed for IT, IP and optical resources allocation and dimensioning, providing fault-tolerant designs for the realistic scenarios defined. Simulation results have been obtained using the Net2Plan NIW (NFV over IP over WDM) open-source framework. Developed algorithms and part of the testing scenarios are available for inspection in public repositories of the EU METRO-HAUL project, the umbrella for our work. Our results show CAPEX benefits in the order of 10% and energy savings in the order of 20-30% stemming from the on-demand resource allocation in the metro. In addition, degree 1 and degree 2 optical nodes have shown to be a sweet spot for applying filterless switching, with mitigated impact of the associated spectrum waste.

Our recently proposed cognitive methodology for optical amplifier gain adjustment, that relies on case-based reasoning, showed optical signal-to-noise ratio improvements over time demonstrating the cognition process regardless the deployed amplifier type. In this paper, we extend our preliminary analysis exploring the cognitive methodology benefits for different and larger network topologies. The obtained results show agreement between networks, demonstrating the methodology suitability regardless the network scenario.

Reconfigurable optical add/drop multiplexers (ROADMs) are key elements in operators’ backbone networks. The breakthrough node concept of architecture on demand (AoD) permits us to design optical nodes with higher flexibility with respect to ROADMs. In this work, we present a five-step algorithm for designing AoD instances according to some given traffic requests, which are able to support subwavelength time switching up to wavelength/superchannel/fiber switching. We evaluate AoD performance in terms of power consumption and number of backplane optical cross-connections. Furthermore, we discuss trade-offs involved in the migration from a fixed to a flexible grid with regard to the optical node size, capacity, and power consumption. We compare several ROADM architectures proposed in the literature with AoD in terms of power consumption and cost. We also study different technologies for enhancing the scalability of AoD. Results show that AoD can bring significant power savings compared to other architectures while offering a throughput of hundreds of terabits per second.

Transport ecosystems that combine Software Defined Networking (SDN) and Network Function Virtualization (NFV) are characterized by an unprecedented network control and resource dynamicity. Manual optimization is unmanageable. In this context, open systems that manage and orchestrate SDN/NFV-enabled networks offer programming frameworks that abstract the low-level particularities in the data-plane forwarding devices and in the hardware appliances that provide the IT resources. Although these open systems present notable complexity, their programming abstractions promote a client layer where third-party applications can provide different functionalities thus enabling Optimization-as-a-Service (OaaS) business opportunities. In this paper, we cover open-source optimization software initiatives for offline planning and online provisioning and orchestration of SDN/NFV networks. With this goal in mind, we first focus on open software (and framework) initiatives through a set of realistic use cases that require optimization in multi-layer optical transport scenarios and ecosystems that combine transport with IT resources. The importance of a joint optimization of both network and IT domains is emphasized, a new paradigm triggered by SDN/NFV technologies. We discuss the theoretical limits to algorithm performances, and review available open-source frameworks for problem modelling that enable the interaction with solvers. Finally we focus on the Net2Plan open-source network planning tool, a Java-based software that suitably embraces the multiple features required in the optimization of joint transport network and IT resource SDN/NFV ecosystems. Recent works based on Net2Plan are reviewed to illustrate its suitability for rapid algorithm prototyping, and for interaction with SDN/NFV-enabled networks.

We present results from the first gridless networking field trial with flexible spectrum switching nodes and 620 km of installed fibre links. Signals at 10G, 12.25G, 42.7G, DP-QPSK 40G, DP-QPSK 100G and 555G are generated, successfully transported and switched using flexible, custom spectrum allocation per channel. Spectrum defragmentation is demonstrated using integrated SOA-MZI wavelength converters. Results show error-free end-to-end performance (BER<1e-9) for the OOK channels and good pre-FEC BER performance with sufficient margin to FEC limit for the 40G and 100G coherent channels as well as for the 555G super-channel.

Metropolitan networks are undergoing a major technological breakthrough leveraging the capabilities of software-defined networking (SDN) and network function virtualization (NFV). NFV permits the deployment of virtualized network functions (VNFs) on commodity hardware appliances which can be combined with SDN flexibility and programmability of the network infrastructure. SDN/NFV-enabled networks require decision-making in two time scales: short-term online resource allocation and mid-to-long term offline planning. In this paper, we first tackle the dimensioning of SDN/NFV-enabled metropolitan networks paying special attention to the role that latency plays in the capacity planning. We focus on a specific use-case: the metropolitan network that covers the Murcia - Alicante Spanish regions. Then, we propose a latency-aware multilayer service-chain allocation (LA-ML-SCA) algorithm to explore a range of maximum latency requirements and their impact on the IT resources for dimensioning the metropolitan network. We observe that design costs increase for low latency requirements as more data center facilities need to be spread to get closer to the network edge, reducing the economies of scale on the IT infrastructure. Subsequently, we review our recent joint computation of multi-site VNF placement and multilayer resource allocation in the deployment of a network service in a metro network. Specifically, a set of subroutines contained in LA-ML-SCA are experimentally validated in a network optimization-as-a-service architecture that assists an Open-Source MANO instance, virtual infrastructure managers and WAN controllers in a metro network test-bed.