Minimizing expected energy in real-time embedded systems

Abstract

We study the problem of minimizing energy consumption in real-time embedded systems that execute variable workloads and are equipped with processors having dynamic voltage scaling (DVS) capabilities. This problem is about how to decide tasks' running speeds (speed schedule) before they are scheduled to execute. In this paper, we show that it is possible to incorporate the dynamic behavior of the tasks into the speed schedule to, along with the dynamic slack reclamation technique, minimize the expected (total) energy consumption in the system.