State Management in Real-Time Cloud and NFV Services

In the context of distributed, real-time Cloud and NFV services, state management is becoming an increasingly important issue that risks impairing the ability of stateful services to provide timely responses in presence of several types of possible failures that might occur at run-time. This paper presents a simulation-based approach for a comparative exploration of the trade-offs in performance vs fault-tolerance and robustness, achievable in distributed end-to-end real-time applications, arising when using different state management and load-balancing strategies.