Research On Multi-user Computing Offloading Strategy In Mobile Edge Computing

The development of the Internet of Things(IoT)and the widespread adoption of 5G communication technology have led to an enormous demand for data and computing resources.Both enterprises and individuals urgently need a technology that can provide powerful computing capabilities and storage capacity.Although cloud computing offers significant computing power and storage space for various applications,its centralized architecture has resulted in issues such as data transmission latency,bandwidth consumption,and energy efficiency.Edge computing,which shifts computing tasks from the cloud to edge devices in the network,has gradually gained attention and been widely applied.Current edge computing research faces numerous challenges,especially with the rapid development of Io T technology.Offloading decisions determine whether to offload computing tasks from resource-constrained devices to edge servers or the cloud,and how to achieve the optimal balance between real-time performance,energy efficiency,and system performance.Designing efficient offloading strategies and allocating resources more reasonably is an urgent problem to be solved.First,this paper introduces the basic concepts and application fields of edge computing,elaborates on the importance and demand of edge computing offloading technology,and discusses its advantages in reducing latency,saving energy,and improving computational efficiency.The challenges faced by current edge computing offloading technology are also described.Secondly,based on multi-user terminal device scenarios,computational models are constructed for different layers in the edge computing architecture,targeting task offloading and transmission scenarios.Task offloading scenarios are modularized without compromising the integrity of the original tasks,dividing the offloading process into several independent subprocesses.For the split subtasks,their dependencies and execution order need to be analyzed.For multi-user terminal device scenarios,specific task offloading scenarios are modularized to meet offloading requirements,reducing offloading overhead and improving system offloading efficiency while completing the computational offloading process.In terms of subtask offloading decisions,Lyapunov optimization theory is used to optimize the offloading process,satisfying the delay requirement of computational offloading while reducing the cost overhead and improving the efficiency and resource utilization of the system during computational offloading.Finally,considering that edge server cluster scenarios are more suitable for practical production applications,research on the offloading strategies under such scenarios is conducted.To ensure overall load balancing and computing capacity,multiple edge server clusters are deployed,with each cluster providing services to terminal users in a specific region.Servers in the cluster can cooperate to process user requests.In this scenario,offloading decisions require higher performance and efficiency in resource management and processing user requests.Optimization of offloading strategies is necessary to improve the energy consumption,latency,and cost evaluation indicators.The improved Coral Reef Optimization algorithm is used to obtain an approximate optimal solution that meets practical needs,resulting in offloading strategies that further improve energy consumption and latency.Experimental results show that the proposed strategy for MEC server cluster scenarios outperforms traditional offloading strategies in terms of latency,energy consumption,and cost.