Energy-aware preemptive scheduling algorithm for sporadic tasks on DVS platform

Dynamic Voltage Scaling (DVS) is a key technique for embedded systems to exploit multiple voltage and frequency levels to reduce energy consumption and to extend battery life. There are many DVS-based algorithms proposed for periodic and aperiodic task models. However, there are few algorithms that support the sporadic task model. Moreover, existing algorithms that support the sporadic model lack of functionalities in terms of energy-saving. In this paper, we propose a novel energy-aware scheduling algorithm named Cycle Conserve Dynamic Voltage Scaling for Sporadic Tasks (CC-DVSST) algorithm which is an improvement to DVSST [1]. There is a large amount of time slack in the DVSST scheduling due to the significant difference between the actual execution time and the worst-case scenario. Introducing DVS with EDF, CC-DVSST scales down the voltage of a processor when tasks are completed earlier than they are expected, so that the slack time can be reused for other tasks, hence saving energy. Experimental results show that CC-DVSST can reduce the total amount of energy consumption up to 46% compared to DVSST while retaining the quality of service by meeting the deadlines. 2012 Elsevier B.V. All rights reserved.