Energy efficient task scheduling for heterogeneous multicore processors in edge computing
Yanchun Liu, Hongyu Qu, Shuang Chen, Xuejun Feng
Abstract
Edge computing faces challenges in energy-efficient task scheduling for heterogeneous multicore processors (HMPs). Existing solutions focus on reactive workload adaptation and energy prediction but fail to effectively integrate dynamic voltage and frequency scaling (DVFS). This paper proposes a novel algorithm integrating task prioritization, core-aware mapping, and predictive DVFS. Our approach outperforms state-of-the-art methods, reducing energy consumption by 20.9% while maintaining a low 2.4% deadline miss rate. Experiments on real HMP platforms demonstrate the algorithm's scalability and adaptability to varying workloads. This work advances energy-efficient edge computing, balancing performance and power constraints.