Run-time Monitoring and Control for Temporal Fault Prevention in Mixed-criticality Systems

Multicore parallelism involve inter-tasks interferences leading to execution timing uncertainties. This is important for safety concerns in industrial applications (automotive for instance). Existing solutions limit significantly the use of available computing resources for low-criticality tasks to keep strong timing guarantees on high-criticality tasks. We propose a run-time monitoring and control approach for mixed-criticality systems to prevent temporal faults. Its objective is to guarantee a high-criticality function end-to-end deadline, even when interfered by low-criticality tasks parallel execution. Such approach aims at allowing better computing resource use with an additional guarantee for a given task chain end-to-end response time. We also propose an experimental protocol to characterise this task chain and calibrate the mechanism. Our approach has been implemented on a Linux-based experimental platform.