AURORAS - Autonomic and disaggregated optical networks leveraging edge computing and photonic technologies

Operators are facing critical issues on their optical transport networks to deploy a wide variety of heterogeneous 5G and IoT services with stringent requirements. This translates into the need to address a required capacity increase (estimated in a factor of 10), to deploy storage and compute resources across the infrastructure, and to conceive autonomic control and management systems for network operation minimizing human intervention, all this with a low cost per user. A recognized strategy for achieving efficiency and cost reduction, in the context of big data centers, relies on disaggregating the software from the hardware, and the use of white boxes with open Application Programming Interfaces (APIs). The AURORAS project aims at extending this concept to the domain of optical networks, where disaggregation provides a new degree of flexibility, at either the optical system or sub-system level, allowing component migration and upgrades without vendor lock-in.

To support this view, AURORAS aims at perform theoretical and experimental research relying on several key pillars. First, the use of Spatial Division Multiplexing (SDM), which has been proposed as the key technology to overcome the capacity crunch that the optical single-mode fibers (SSMFs) are facing, along with the integration of suitable photonic technologies and devices in novel transmission systems able to efficiently exploit the multiple dimensions at lower cost. Next, the systematic deployment of distributed computing and storage resources bound to transmission and switching nodes, the latter with partial or full disaggregation, across the underlying infrastructure for maximum service deployment flexibility. Then, the design of hybrid control plane architectures extending Software Defined Networking (SDN), using systematic and uniform model-driven development and data modelling, that supports autonomic networking and local control loops. This leverages advances in cognition and automatic learning techniques, both at the transmission and network operation level, which require efficient monitoring, telemetry and data analytics. Finally, the conception of an extended NFV-based service platform, with state of art service deployment and lifetime management, while supporting network slicing and sharing that characterize modern transport networks.

AURORAS work plan and methodology is organized in five work packages, covering these topics. As with the previous projects of the group, AURORAS has a high experimental component, including the development of hardware and software prototypes, their integration and experimental performance evaluation in our ADRENALINE testbed. AURORAS’ goals are relevant to and aligned with not only Spanish “Reto 7”, but also major R&D strategies including H2020 ICT as well as the upcoming Framework Program. Although the Metro segment is the main reference scope of AURORAs, it is expected that major outcomes may also be applied in other network segments.

Finally, AURORAS aims at having a scientific, social and economic impact by disseminating main results and a commitment to perform technology transfer to interested potential customers. Our dissemination strategy includes, but is not limited to, high-quality scientific publications, combining the expertise of two complementary research areas, optical transmission and subsystems along with network control and service management.