Performance and Power Consumption of Real-Time Task Scheduling for Embedded Systems Devices

Real-Time Operating Systems (RTOS) play a pivotal role in the embedded systems domain by ensuring timely and predictable task execution. This characteristic is crucial for applications where delayed or missed deadlines can lead to system failure or degraded performance. Therefore, this aspect must be explored for constrained devices in Internet of Things (IoT) applications. This study analyzes three ready queue algorithms and varying CPU frequencies and low-power states in Zephyr RTOS, focusing on how prioritizing critical tasks impacts performance and energy consumption. Our results indicate that evaluating CPU task scheduling performance and power consumption provides valuable insights for enhancing energy savings in IoT applications. Optimizing active power consumption improves energy savings by completing tasks faster at higher frequencies and switching to a deep sleep state.