Optimization Of Load Balancing Technology Based On Adaptive BP Neural Network Improved Particle Swarm Optimization Algorithm

With the rapid development of the Internet,the number of Web server-side visits is also on the rise.Web server-side traffic increased,while the performance of the server has higher requirements.Server cluster technology is a common method to solve this problem.This method improves the performance of the cluster system through the allocation of requests,and achieves the purpose of load balancing.This article describes the methods and techniques commonly used in load balancing systems.Some load balancing algorithms don't consider dynamic changes in performance,at the same time,the improved Particle Swarm Optimization algorithm is used to optimize the load balancing schedule.In this paper,a load balancing strategy based on self-adaption BP neural network and particle swarm optimization is proposed.At the same time,this new strategy is used to test the load balancing system.The specific research contents are as follows:(1)Load balancing technology researchThis article first outlines the advantages of load balancing techniques,and then analyzes the concepts and working principles of load balancing techniques.Based on the advantages and principles of load balancing technology,the goal of load balancing in the whole cluster system is analyzed.According to the classification of load balancing strategy,this paper focuses on the dynamic load balancing algorithm and static load balancing algorithm.Finally,the load balancing technology is compared and analyzed from the aspects of system performance,response time,algorithm complexity and development cost.(2)Research on dynamic load balancing scheduling strategy based on improved particle swarm optimizationIn order to optimize the load balancing technology and improve its response efficiency,this paper first analyzes the basic model of load balancing and studies the basic dynamic load balancing method and load vector selection and acquisition.At the same time,a load balancing scheduling algorithm based on adaptive BP neural network is proposed.Based on adaptive BP neural network faults,add the improved particle swarm algorithm to tune it and studied both the combination load balance.Finally,the theoretical analysis and simulation results verify that the proposed combination load balancing scheduling strategy can better improve the system response time than the existing load balancing strategy.In order to verify the stability and reliability of the combined algorithm,this paper briefly introduces the design flow and implementation of load balancing system.In addition,the load balancing test system environment was built.Simulation and test results show that the proposed self-adaption BP neural network-Particle Swarm Optimization algorithm for the system to request the corresponding time optimization effect is obvious.Therefore,the algorithm studied in this paper has a certain theoretical reference value and practical value.