【摘要】 针对系统不能动态反映服务器的实时负载情况的缺陷,研究常见的传统负载均衡算法,同时引入一种服务器集群OPNET,以解决多用户并发访问的问题,并对传统负载算法进行优缺点对比。从而引入BP神经网络反馈机制,提出一种基于神经网络反馈机制的改进型负载均衡算法。通过对常见的传统负载均衡算法和改进型的负载均衡算法进行仿真对比,结果表明,改进型的算法在服务器、CPU利用率以及任务请求速率运行上有较大的改善。更多还原

【Abstract】 In view of the fact that the system cannot dynamically reflect the real-time load situation of the server, studies the common traditional load balancing algorithms, and introduces the OPNET cluster, in order to solve the problem of multi-user access, of the traditional algorithm,and compares the advantages and disadvantages of load. Introduces the feedback mechanism of BP neural network, and proposes an improved load balancing algorithm based on neural network feedback mechanism. Through the simulation of conventional load balancing algorithm and improved load balancing algorithm and the result shows that the improved load balancing algorithm has a significant improvement in server, CPU utilization and task request rate.更多还原