PANTAS: An energy-aware task scheduling and offloading method in edge-cloud computing using metaheuristic algorithms

The increasing complexity of mobile applications necessitates substantial computational power, driven by rapid technological advancements. However, progress in battery technology has lagged behind, restricting the ability of mobile devices to execute tasks locally. To address this limitation, task offloading, which delegates computational tasks to fog or cloud servers, has emerged as a critical solution. This research proposes a parallel simulation annealing with time arrival scheduling (PANTAS) algorithm to optimize two key decisions: selecting which tasks to offload and determining the optimal timing for offloading. Extensive simulations demonstrate that the PANTAS algorithm improves up to 10% energy efficiency with 9% lower makespan or 9% higher reliability, outperforming the other five baseline algorithms in energy consumption while minimizing makespan and maximizing reliability.