Research on Dynamic Load Balancing Based on Artificial Bee Colony Algorithm

With the rapid development of the Internet, explosive access requests pose significant challenges to server performance and capacity. Considering the inefficiencies of current load balancing algorithms, such as decreased system efficiency at lower loads and suboptimal allocation efficiency at higher loads, this paper proposes a dynamic load balancing algorithm based on the Nginx server and the artificial bee colony algorithm. Through the artificial bee colony algorithm, the threshold for the conversion between static and dynamic load balancing is calculated. If the server load is below this threshold, a static algorithm is employed; otherwise, a dynamic algorithm is used. This algorithm utilizes server performance parameters such as CPU performance, memory performance, disk I/O, and network bandwidth as evaluation metrics. Through comparative experiments, the algorithm’s effectiveness is demonstrated in experimental environments, showing superior load balancing effects compared to weighted round-robin and least connection algorithms, effectively improving average server response time and actual concurrent connection numbers.