Dynamic I/O power management in real-time systems with multiple-state I/O devices

Power consumption is now an important design parameter for distributed sensor networks. While dynamic (or software-controlled) power management (DPM) is being increasingly used as an alternative to inflexible hardware solutions, DPM for hard real-time systems has received relatively little attention. In particular, energy-driven I/O device scheduling for real-time systems has received even less attention. We present an online DPM algorithm, which we call multi-state constrained low energy scheduler (MUSCLES), for hard real-time systems. MUSCLES generates a sequence of power states for each multiple-state I/O device while guaranteeing that real-time constraints are not violated. It also minimizes the energy consumed by the I/O devices that are used by a task set. MUSCLES is energy-optimal under the constraint that the start times of the tasks are fixed. We present several realistic case studies to show that MUSCLES reduces energy consumption significantly for real-time systems.