Probabilities and Mixed-Criticalities: the Probabilistic C-Space

Probability distributions bring flexibility as well as accuracy in modeling and analyzing real-time systems. On the other end, the adding of probabilities increases the complexity of the scheduling problem, especially in case of mixed-criticalities where tasks of different criticalities have to be taken into account on the same computing platform. In this work we explore the flexibility of probabilistic distributions applied to mixed-critical task sets for defining the probabilistic space of Worst Case Execution Time and evaluating the effects of changes on the task execution conditions. Finally, we start formalizing and making use of probabilistic sensitivity analysis for evaluating mixed-critical scheduling performance.