2017 IEEE 23rd International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA) (2017)
Aug. 16, 2017 to Aug. 18, 2017
Alessandra Melani , Scuola Superiore Sant'Anna, Pisa, Italy
Renato Mancuso , University of Illinois at Urbana-Champaign, USA
Marco Caccamo , University of Illinois at Urbana-Champaign, USA
Giorgio Buttazzo , Scuola Superiore Sant'Anna, Pisa, Italy
Johannes Freitag , Airbus Group Innovations, Munich, Germany
Sascha Uhrig , Airbus Group Innovations, Munich, Germany
Although multicore chips are quickly replacing uniprocessor ones, safety-critical embedded systems are still developed using single processor architecture. The reasons mainly concern predictability and certification issues. This paper proposes a scheduling framework for handling Integrated Modular Avionics (IMA) on multicore platforms providing predictability as well as flexibility in managing dynamic load conditions and unexpected temporal misbehaviors of multicore. A new computational model is proposed to allow specifying a higher degree of flexibility and minimum performance requirements. Schedulability analysis is derived for providing off-line guarantees of real-time constraints in worst-case scenarios, and an efficient reclaiming mechanism is proposed to improve the average-case performance. Simulation and experimental results are reported to validate the proposed approach.
Multicore processing, Aerospace electronics, Servers, Switches, Real-time systems, Interference
A. Melani, R. Mancuso, M. Caccamo, G. Buttazzo, J. Freitag and S. Uhrig, "A scheduling framework for handling integrated modular avionic systems on multicore platforms," 2017 IEEE 23rd International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA), Hsinchu, Taiwan, 2017, pp. 1-10.