Adaptive Seamless Redundancy to Achieve Highly Dependable MQTT Communication

Nowadays, large enterprises, critical infras- tructures, and utilities are complex systems, and consist of a number of subsystems, often located far away from each other, that coordinate their operations over geographic net- works. Typically, they are interconnected through the Inter- net and communicate by means of transport control proto- col (TCP)/IP-based protocols. When system-level resilience is required, extremely high dependability is demanded from the network. In this article a solution is described and evalu- ated that loosely resembles SDN architectures. A redundant version of MQTT is used for the data plane, whereas an adaptive mechanism is implemented in the control plane that evaluates the quality of paths and dynamically selects the best choice. To maximize dependability, yet keeping resource consumption to acceptable levels, concepts bor- rowed from reinforcement learning are exploited. An ex- perimental campaign corroborated our expectations and showed the practical feasibility of our proposal.