Achieving Resilience in Software-Defined Networks

Establish architectures and mechanisms that ensure resilience of software-defined networks, including high availability of the logically centralized control plane and robust data-plane fault tolerance, to address the currently unresolved challenge of SDN resiliency.

Background

Within the discussion on centralized versus distributed controller designs, the authors highlight single points of failure and the need for fault tolerance. Distributed controllers can improve availability and scalability but introduce trade-offs in consistency and coordination.

The paper points readers to a dedicated resilience section and explicitly notes that end-to-end resilience—spanning both control and data planes—remains an open challenge, underscoring the need for principled designs and mechanisms for fast failure detection, protection, and recovery.

References

SDN resiliency as a whole is an open challenge that will be further discussed in Section \ref{sec:resiliency}.

— Software-Defined Networking: A Comprehensive Survey  (1406.0440 - Kreutz et al., 2014) in Section IV: Network Operating Systems / Controllers (Centralized vs. Distributed)

Finally, the current design assumes trustworthy telemetry and well-behaved participants; the effects of telemetry inaccuracies, controller failures, and adversarial clients remain open directions for future work.

— HybridFLow: SDN-Orchestrated Client Partitioning for Hybrid Federated Learning  (2609.10404 - Hamdan et al., 9 Sep 2026) in Section 5, Limitations of the Study

Persisted datastores let a vehicle boot and operate with an unreachable controller, so management-plane availability is not on the critical path; controller redundancy, an explicit fail-safe policy, and the degradation logic that must gate revoking a safety-relevant actuator remain open, as it belongs to SDN resilience concepts.

— Lightweight Zero Trust via Automotive SDN  (2609.09817 - Wiemer et al., 9 Sep 2026) in Section 4, Conclusion, paragraph 'Limitations' (L5)