Abstract: Fault-tolerance Energy-saving Scheduling Algorithm Based on Checkpoint Scheme (CPFTES) is proposed that considers fault tolerant and power management in the real-time system with processors providing the discrete frequency and voltage. Fault tolerance is achieved via optimal checkpoints scheme and power management is carried out using dynamic voltage scaling. The optimal checkpoints can help the task to guarantee the timing constraints and reliability. If a task fails, it will restart from a saved checkpoint and execute at maximum speed to guarantee that the deadlines are met. H_SPEED and A_SPEED are proposed which are the two speed selection strategies of the algorithm. Simulation results show that CPFTES_A_SPEED provides about 66.18% of energy savings compare to CPFTES_NODVS and provides about 0~19.41% of energy savings compare to CPFTES_H_SPEED under PXA250 processors.

Key words: Computer System Architecture, Fault-tolerance, Checkpoint, Dynamic Voltage Scaling, power management