ABSTRACT This paper presents a novel framework named Optimal Edge Server Placement and Offloading in MEC (OESPO‐MEC). The OESPO‐MEC model features two primary components: computational offloading and resource allocation. The core objective of computational offloading is to move tasks from User Equipment (UE) to edge servers, thereby reducing data travel distance, cutting processing time, and achieving lower latency and quicker application responses. However, this approach can lead to challenges such as performance degradation and data loss due to discrepancies between computational demands and available server resources. To overcome these challenges, we propose the LAOO algorithm, which optimally places edge servers and manages tasks across base stations. The Lyrebird Assisted Osprey Optimization algorithm (LAOO) is designed to meet client needs by minimizing time delays, lowering energy consumption, and balancing load variance. Additionally, the model integrates Multi‐Head Attention‐based SqueezeNet (MHA‐SqN) model for sophisticated task offloading decisions, which determine whether tasks should be processed locally on UE or offloaded to the edge server. Once offloading decisions are made, the LAOO method is utilized for optimal resource allocation on Virtual Machines (VMs), considering factors including execution time, cost, task priority, and load imbalance. Moreover, the proposed LAOO strategy is competing with traditional algorithms including BES, PSO, GWO‐WOA, EHO, OOA, and LOA in terms of various comparative analyses. As a result, the LAOO scheme has achieved a minimal cost rate of 0.1463 at the 25th iteration, demonstrating faster convergence and showing outstanding performance in balancing load distribution across edge servers compared to the conventional methods.
Kurni et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: