ELITE: Energy and Latency-Optimized Task Offloading for DVFS-Enabled Resource-Constrained Devices in MEC

Multi-access Edge Computing (MEC) technology offers promising support for modern, computation-intensive, and time-sensitive applications. Many of these applications are generated by resource-constrained handheld or mobile UE. Due to limited resources, offloading certain parts of these applications (tasks) to connected MEC servers becomes essential. However, MEC servers also have limited resources compared to cloud servers, highlighting the need for efficient task offloading policies for UE devices and optimal resource allocation policies for MEC servers. This paper introduces ELITE (Energy and Latency-optimized Task Offloading and Resource Allocation for DVFS-Enabled devices), a novel solution to the energy and latency minimization problem in a cooperative heterogeneous MEC architecture. The proposed policy aims to minimize the energy consumption of the UE devices and the latency of the applications while satisfying application deadlines and dependency constraints. Furthermore, we consider the UEs to be enabled by dynamic voltage and frequency scaling (DVFS). Through extensive simulations using a real dataset, we demonstrate that our proposed strategy surpasses the state-of-the-art policy, achieving a remarkable 10% reduction in latency and an impressive 2 \(\times \) reduction in energy consumption of UE devices.