ARGENT: Energy-Aware Scheduling in Edge Computing Using Energy Valley Optimizer

This paper presents a novel approach for energy-aware task scheduling in heterogeneous edge computing environments. The Energy Valley Optimizer (EVO) is exploited for the optimal allocation of tasks to edge devices with limited resources. The primary objective of this algorithm is to minimize energy consumption and task completion time while achieving optimal resource utilization. The EVO updates task positions in the search space using our proposed alpha, beta, and gamma reduction processes, avoiding local optima. Our evaluations on both small and large scales with real-world datasets indicate that our approach reduces energy consumption by an average of 14.42 %, reaching up to 20.96% savings, and decreases task completion time by an average of 15.98%, with a peak improvement of 24.42%, in comparison to state-of-the-art methods. Additionally, EVO shows a 5.28% improvement in success ratio, which means it is robust in meeting task deadlines.