Model-based optimization of ARINC-653 partition scheduling

Abstract The architecture of ARINC-653 partitioned scheduling has been widely applied to avionics systems owing its robust temporal isolation among applications. However, this partitioning mechanism causes the problem how optimize partition a complex system while guaranteeing schedulability. In paper, model-based optimization approach is proposed. We formulate as parameter sweep application, which searches for optimal parameters with respect minimum processor occupancy via an evolutionary algorithm. An modeled set timed automata in model checker UPPAAL. optimizer tentatively assigns settings models and subsequently invokes UPPAAL verify schedulability well evaluate promising solutions. space explored algorithm that combines refined genetic operators self-adaptation evolution strategies. experimental results show applicability our method.