ABSTRACT
Tolerating faults and minimising energy consumption in embedded systems is a difficult task due to the fact that the two objectives are antagonistic. In this paper, we propose a new approach based on graceful degradation to reduce jitter of battery life and thereby energy consumption in fault-tolerant embedded systems. In case of faults, the affected task is re-executed. In our solution, the energy level of battery is periodically verified, and if we detect that the continuity with the current operating mode leads to jitter, the system gracefully degrades to the adequate operating mode. In such degraded mode, the dynamic voltage scaling technique is used to save energy. The effectiveness of using graceful degradation is depending on the application criticality level. Simulation results show that the use of graceful degradation can reduce jitter of battery life, and thereby can minimise energy consumption.
Disclosure statement
No potential conflict of interest was reported by the authors.
Additional information
Notes on contributors
Salim Kalla
Salim Kalla received his Master's in Computer Science from the University of Batna. He is currently an Assistant Professor and PhD candidate with Computer Science Department in the same University. His research topics are scheduling and fault tolerance for critical real-time and distributed systems.
Riadh Hocine
Riadh Hocine received his PhD in Computer Science from the University of Batna. He is currently an Assistant Professor with Computer Science Department in the same University. His special fields of interest include dependability, distributed and real time systems.
Hamoudi Kalla
Hamoudi Kalla received his PhD in Computer Science from the Polytechnic Institute of Grenoble, France, in 2004. He is currently a Lecturer in Computer Science Department of Batna 2 University, Algeria, since 2006. His research focuses on power and fault tolerance scheduling in realtime embedded systems.
Abderrezak Chouki
Abderrezak Chouki received his Master's in Computer Science from the University of Hadj Lakhdar Batna. His special fields of interest include scheduling algorithms, reliability and fault tolerance in real-time distributed embedded systems.