大型云计算多服务器环境下故障节点定位仿真  

A Large Cloud Computing Fault Node Localization Multi- server Environment Simulation

作　　者：谭营军[1] 张铁头[1] 李翠霞[2] 

机构地区：[1]河南职业技术学院信息工程系,郑州450046 [2]郑州大学软件技术学院,郑州450002

出　　处：《控制工程》2014年第6期909-912,共4页Control Engineering of China

基　　金：河南省高等教育教学改革研究项目(2014SJGLX381)

摘　　要：云计算环境下多服务器故障节点准确定位关系到设备的安全。在云计算环境中,大量不同编码规则的服务器同时工作,单个服务器形成服务器节点,由于编码规则的不同,服务器节点之间存在较强的差异性,传统的服务器故障节点定位模型都是以相同规则编码下的服务器统一特征对比为基础,完成故障定位的,一旦服务器之间是不同规则的,定位需要进行区分,定位效果不好。提出基于粒子群单循环搜索算法的云计算中多服务器故障节点定位方法。提取云计算网络中不同服务器故障节点特征,计算对应特征的权重,并对特征提取误差进行补偿,实现差异化故障节点特征提取。根据上述特征,构造初始粒子群,计算粒子的更新速度和空间位置,搜索故障特征最优解,得到的即是故障节点。实验结果表明,利用改进算法进行云计算中故障节点定位,能够极大的提高定位精度,在服务器数量较多,服务器节点之间差异较大的情况下,这种优势更加明显。More accurate positioning server failure node cloud computing environment is related to the safety of the equipment. In a cloud computing environment, a large number of different encoding rules of the server work at the same time, the formation of a single server server nodes, because of the different encoding rules, server nodes, and strong differences between the traditional server failure node positioning model is based on the same server unified under the rules of coding feature comparison, on the basis of complete fault location, once the server is different rules, positioning to distinguish, positioning result is bad. Based on particle swarm single search algorithm in cloud computing server failure node positioning method. To extract different servers in the cloud computing network fault node, to calculate weight of the corresponding characteristics, and the feature extraction of error compensation, to differentiate the fault feature extraction. According to the above characteristics, constructs the initial particle swarm, calculate the renewal speed of particles and spatial location, search the optimal solution, the fault features is fault node. Experimental results show that the improved algorithm in cloud computing fault node localization, can greatly improve the positioning accuracy, the quantity is more, the server server nodes between differences, this advantage is more obvious.

关 键 词：云计算 多服务器 节点定位 故障 

分 类 号：F127[经济管理—世界经济]