Robust Partitioned Scheduling for Static-Priority Real-Time Multiprocessor Systems with Shared Resources

International audienceWe focus on the partitioned scheduling of sporadic real-time tasks with constrained deadlines. The scheduling policy on each processor is static-priority. The considered tasks are not independent and the consistency of these shared data is ensured by a multiprocessor synchronization protocol. Considering these assumptions, we propose a partitioned scheduling algorithm which tends to maximize the robustness of the tasks to the Worst Case Execution Time (WCET) overruns faults. We describe the context of the problem and we outline our solution based on simulated annealing