云计算环境下网络节点负荷检测算法

Network Node Load Detection Algorithm in Cloud Computing Environment

在云计算环境下网络节点负荷检测过程中,由于云计算环境下网络节点分布的发散,节点之间特征不同,使得检测过程中出现特征与噪声的干扰,传统的检测算法,在进行检测时忽略了节点特征和去噪对检测结果的重要性,导致传统的方法精确度低,实用性差。提出改进遗传算法的云计算最大压力点检测方法。借助系统的监测设备提取最大流量负荷压力点的特征信号,采用非线性滤波方法对最大流量负荷压力点特征参数进行去噪,同时将其离散化,并计算最大流量负荷压力点的负载系数,融合遗传算法,并引入精英选择策略,将结果与系统负载的最高系数阈值进行比较,有效的实现了最大流量负荷压力点的检测。实验仿真证明了改进方法的精确度高,实用性强。

A detection method of the maximum pressure point in cloud computing environment is proposed based on improved genetic algorithm. The characteristic signal of the maximum flow load pressure point is extracted with the help of monitoring equipment, the characteristic parameters of the maximum flow load pressure point are denoised by adopting the nonlinear filtering, and the load coefficient of the maximum flow load pressure point is calculated. By in- tegrating genetic algorithm, introducing elite selection strategy, and comparing with the highest coefficient of system load threshold, the detection of the maximum flow load pressure point is effectively implemented. Simulation result proves that the improved method has high precision and strong practicability.