Accurate Strategy for Mixed Criticality Scheduling.

In mixed criticality systems, functionalities with different criticalities share the same execution platform. The goal of a mixed criticality scheduling algorithm is to ensure a safe execution of the highest criticality tasks while using efficiently the execution platform. Classical real-time scheduling algorithms for mixed criticality systems propose to stop the execution of low criticality tasks to ensure that the highest criticality tasks meet their deadlines. In this work, we propose the Accurate Fault Mode (AFM) strategy for the scheduling problem of mixed criticality real-time systems. The advantages of this strategy are that firstly it can reduce the number of stopped low criticality tasks and secondly, that it allows the designer to define the low criticality tasks to be stopped depending on the criticality configuration of the system. Using model checking for timed game automata, we propose an exact feasibility test and exact schedulability tests for fixed priority and earliest deadline first scheduling algorithms for AFM strategy.