Systemwide Energy Minimization in Real-Time Embedded Systems

Traditionally, dynamic voltage scaling (DVS) techniques have focused on minimizing the processor power consumption as opposed to the entire system energy. However, the slowdown resulting from DVS can increase the energy consumption of components like memory and network interfaces. Furthermore, leakage power consumption, which is increasing with the scaling device technology, must also be considered. In this paper, we present an algorithm to compute task slowdown factors based on the contribution of the processor leakage and standby energy consumption of the resources in the system. We combine slowdown with procrastination of task executions to extend sleep intervals which significantly reduces the leakage energy consumption. We show that the scheduling approach minimizes the total static and dynamic energy consumption of the systemwide resources. In this work, we consider a real-time system model using the Earliest Deadline First (EDF) policy. Our simulation experiments using randomly generated task sets show on an average 10% energy gains over traditional dynamic voltage scaling. Our procrastination scheme increases the average energy savings to 15%.