A Novel Priority Based Scheduler for Asymmetric Multi-core Edge Computing

Abstract

Asymmetric multi-core processors require careful assignment of tasks to the appropriate cores. Most of the edge computing devices are operated on Linux. However, the current Linux scheduler assigns tasks to cores without considering their capabilities. This can lead to high-priority tasks being assigned to energy-efficient cores while low-priority tasks are assigned to high-performance cores. This results in a decrease in the overall system performance. A new algorithm has been proposed to address this issue that considers the core’s capabilities and the task’s priority. This algorithm requires prior knowledge of the core’s speed due to the asymmetric nature of the cores. The priorities of the tasks are fetched, and then tasks are divided into high, medium, and low classes. High-priority tasks are scheduled on high-performance cores, while medium and low-priority tasks are scheduled on energy-efficient cores. The proposed algorithm performs better than the existing Linux task scheduling algorithm for high-priority tasks, improving task scheduling time by up to 16%. © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.