Increasing Resilience of SD-WAN by Distributing the Control Plane

Modern WAN interconnects utilize SD-WAN to automatically respond to network changes and improve link utilization, latency, and availability. Therefore, they incorporate some kind of centralized controller that collects network state from all managed gateways, calculates suitable forwarding actions, and distributes them accordingly. However, this limits the robustness and availability of the network control plane, especially in the event of node or partial network outages. In this paper, we propose a distributed and highly robust SD-WAN control plane without any central or regional controller. Our solution can handle arbitrary device failures as well as network partitioning. The distributed forwarding decisions are based on user-defined, dynamically evaluated path cost functions, and consider not only path quality but also quality fluctuations. The evaluation shows that our approach can handle several thousand SD-WAN gateways and multiple hundred network policies in terms of computation. Yet, it also highlights that distributing all decisions requires additional communication bandwidth, which may limit the number of supported connections in certain scenarios.