Meta-scheduling techniques for energy-efficient robust and adaptive time-triggered systems

摘要

Many real-time embedded systems need to adapt to run-time events such as faults or changing environmental conditions. In time-triggered systems, which are prevalent in safety-critical applications, adaptivity can be realized by switching between static schedules, which are computed by scenario-based meta-scheduling techniques and tools. These tools exploit metadata and meta-scheduling models and algorithms. An example of an important event type is dynamic slack, which can be used to improve the system's energy efficiency by changing communication and execution plans. Likewise, the reaction to fault events involves the passivation of faulty components and the migration to spare resources. These mechanisms improve the robustness of time-triggered real-time systems and provide high potential for time and energy management. Nowadays, energy saving methods and algorithms are used in a wide-range of applications (e.g., mobiles phone), while their applicability in safety-critical systems is still limited. In this paper, we present meta-scheduling techniques and a meta-scheduler tool (MeS) to generate meta-schedules for time-triggered systems. We also experimentally evaluate the energy efficiency and timing based on slack events and fault events. Finally, results are visualized and validated with a meta-scheduling visualization tool (MeSViz).