Abstract
In this paper, a novel resource allocation approach dedicated to hard real-time systems with distinctive operational modes is proposed. The aim of this approach is to reduce the energy dissipation of the computing cores by either powering them off or switching them into energy-saving states while still guaranteeing to meet all timing constraints. The approach is illustrated with two industrial applications, an engine control management and an engine control unit. Moreover, the amount of data to be migrated during the mode change is minimised. Since the number of processing cores and their energy dissipation are often negatively correlated with the amount of data to be migrated during the mode change, there is some trade-off between these values, which is also analysed in this paper.
Highlights
Electronic control units (ECUs) have become key components of contemporary cars and compose powertrain, safety and comfort subsystems
The contribution of that paper has focused mainly on the issue of schedulability in each mode and during mode changes, whereas in this paper we present multiple solutions in a form of a Pareto frontier to choose a solution representing a trade-off between migration time and the energy dissipated in the future mode
6 Conclusions An approach for runnable migration in a Network on Chip (NoC)-based multi-core system has been proposed as a way to decrease the number of cores needed for guaranteeing safe execution of a hard real-time software
Summary
Electronic control units (ECUs) have become key components of contemporary cars and compose powertrain, safety and comfort subsystems. Throttle is wide open and the number of RPM is larger than the idle RPM level It has been stressed in [7] that execution times of particular runnables may differ significantly for various modes of an ECU and applying different mappings for each operating mode may be beneficial. It means that in some particular modes less computational resources are required to execute all runnables before their deadlines These extra computational cores can be either switched off, or transferred into a more energy saving state if they support dynamic voltage and frequency scaling (DVFS) technique. To the best of our knowledge, there has been no proposal of any other method that jointly addresses the problems of (i) guaranteeing no hard deadline violation during mode switching, (ii) performing schedulability analysis for NoC-based multi-core systems, (iii) finding a trade-off between migration data amount and energy dissipation. Deadlines for mode switching time between each neighbouring pair of modes shall be provided
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
