Applying dynamic priority scheduling scheme to static systems of pinwheel task model in power-aware scheduling

Area of Science:

• Computer Science
• Real-Time Systems
• Embedded Systems

Background:

• Power-aware scheduling is crucial for reducing energy consumption in hard real-time systems using dynamic voltage scaling (DVS).
• The pinwheel task model, characterized by static priorities and predictable execution, is widely applicable to embedded and ubiquitous computing.

Purpose of the Study:

• To demonstrate the effectiveness of dynamic priority scheduling for power-aware optimization within the pinwheel task model.
• To introduce a novel algorithm that enhances energy savings compared to existing static methods.

Main Methods:

• Application of dynamic priority scheduling to the pinwheel task model.
• Development of a novel power-aware scheduling algorithm leveraging slack exploitation under preemptive earliest-deadline first (EDF) scheduling.
• Algorithmic complexity analysis (O(n)).

Main Results:

• Dynamic priority scheduling proves more effective in energy saving than static priority scheduling for the pinwheel task model.
• The proposed algorithm reduces energy consumption by 10-80% compared to existing algorithms.
• The method is tractable and applicable to small-sized embedded systems.

Conclusions:

• Dynamic priority scheduling is a viable and superior approach for power-aware scheduling in static pinwheel task systems.
• The novel algorithm offers significant energy reduction while maintaining system tractability.
• This research contributes to more energy-efficient embedded and ubiquitous computing.