Handing over highly demanding tasks to remote or nearby computing units helps accommodate the service Quality of Service (QoS) requirements, and compensates for the limited computational capabilities of User Equipment (UE) such as smartphones and tablets. Task offloading is a promising technique being proposed for Virtualized Edge (VE) environments to solve a wide range of issues, frequently with the aim of enabling resource-intensive low-latency services. However, the volatile nature of 5G and B5G networks, as they continuously change due to dynamic policies, optimization processes, and users’ mobility, formalizes a major obstacle facing offloading and overall resource orchestration. To cope with such a challenge, under the scope of Multi-access Edge Computing (MEC), a three-tier fuzzy-based orchestration strategy is proposed with the aim of offloading the users’ workload to the optimum computing units to support stricter QoS requirements and reduce the perceived service delay. To evaluate our solution, we compare the proposed workload orchestrator with different employed algorithms. The evaluation shows that our orchestrator achieves nearly ideal performance, and outperforms the state-of-the-art approaches considered. © 2021 IEEE.
Add the full text or supplementary notes for the publication here using Markdown formatting.