First experimental assessment of ABNO-driven in-operation flexgrid network re-optimization

Traffic affected by link failures can be recovered using path restoration schemes. In dynamically operated networks provided with a control plane, restoration algorithms run in a centralized element, such as the path computation element (PCE). To increase traffic restorability in flexgrid networks, multiple paths, subconnections , can be used to restore every single affected connection. However, the multipath restoration scheme might result in a poor resource utilization entailing a lesser grade of service. In-operation network planning algorithms can be used to mitigate this problem once the failed link is repaired; we propose solving the so-called multipath after failure repair optimization problem (MP-AFRO) to reduce subconnections count by aggregating those belonging to the same original connection and rerouting the resulting connection to release spectral resources. The MP-AFRO problem is modeled using a mixed integer linear program formulation. In view of the complexity of the model and the limited time to solve the problem, we propose a heuristic algorithm that provides a good tradeoff between complexity and optimality. The performance on the MP-AFRO heuristic is firstly validated by simulation. Next, the heuristic algorithm is deployed inside an in-operation planning tool in the form of back-end PCE (bPCE) inside the application-based network operations architecture controlling a network; the bPCE is connected to the centralized active stateful PCE. MP-AFRO is experimentally demonstrated using a distributed field trial test-bed connecting the premises of Telefonica (Madrid), CNIT (Pisa), and UPC (Barcelona).