Low-latency controller load balancing strategy and offloading decision generation algorithm based on lyapunov optimization in SDN mobile edge computing environment

To solve the problem of multi-SDN controller load balancing, a low-latency controller load balancing switch migration algorithm is proposed, and a load balancing framework consisting of three modules of load monitoring, decision-making, and switch migration is designed. Migrate the switch with the highest request rate to the controller with stronger processing capacity and closer distance, and achieve load balancing through multiple iterations, effectively solving the problem that the static controller deployment scheme cannot cope with the dynamic network environment. Since the current research on computing offloading does not consider the stability of the MEC system, an offloading decision-generation algorithm based on Lyapunov optimization is proposed. This algorithm designs a task queue scheduling model to transform the system stability problem into a queue backlog problem, considering the profit of the edge server and the delay of task processing, and establishing a resource optimization model to maximize the profit of the MEC system under the premise of meeting the stability and delay requirements of the MEC system. The experimental results show that the proposed controller load-balancing algorithm can speed up the load-balancing process and reduce the average response delay of the system by about 22.1% while maintaining high throughput. The proposed computing offload algorithm can reduce the average delay of the system by 52%, better allocate computing tasks, and make the edge server obtain higher profits.

Automated summary

To address the issue of load balancing in multi-SDN controllers, a low-latency controller load balancing switch migration algorithm is proposed. It consists of a load monitoring module, a decision-making module, and a switch migration module. By migrating the switch with the highest load to a controller with better processing capacity and shorter distance, load balancing is achieved. Additionally, an offloading decision-generation algorithm based on Lyapunov optimization is proposed to consider the stability of the MEC system and maximize its profit while meeting stability and delay requirements.