企业的网络故障定位系统研究与设计

发布时间：2018-04-25 02:12

本文选题：蚁群算法 + 信息素　；参考：《武汉轻工大学》2016年硕士论文

【摘要】：本文选题来源湖北省自然科学基金资助项目(编号2009Chb008)—流控制传输协议的拥塞控制新型计算模型研究,其对象是网络通信必须的物理设备。传统的网络管理是监测管理对象的故障现象,或接受管理对象的故障报告,再创建和维护故障日志记录库,并人工对故障日志进行分析实现网络故障定位,而现在的大型网络中网络业务复杂,各类业务应用导致的网络故障原因有很多的不确定性,这种不确定性使人工定位网络故障带来了更加严峻的挑战。蚁群算法是众多不确定性技术的一种,有类似TSP、VRP模式的思想,涉及到网络搜索问题,有良好的正反馈性和容错性,在复杂性和精确性两个方面也有很好的平衡,适用于大型网络环境,所以本文以蚁群算法定位故障的精准性、收敛性和效率性为目标,深入研究与分析了大型企业的网络故障定位技术。本文首先深入总结了常用的网络故障定位技术,分析了确定性技术和不确定性技术的优缺点,提出了一种基于蚁群算法的网络故障定位方法。在大型企业网络环境中引入了网络分析器和网络探针,并改进了它们之间的通信协议。为了适应故障的动态性,引入了时间片信息;为了解决网络故障上下游问题,引入了信息素浓度。此外还设计了大型企业网络故障定位系统模型,并通过实验仿真,验证了在网络节点较多、部署网络探针最少和故障搜索次数最小的情况下本算法的正确性。最后,本文以大型企业(湖北广电火凤广告电视媒资库(备播)系统)的网络环境为例,对该企业网络的安全需求,进行了全面分析与设计,通过仿真平台GNS3对该企业内部的网络进行搭建,并运用基本的网络技术如VLAN技术、HSRP技术、STP技术、TRUNK技术、链路聚合技术、防火墙技术、冗余技术、VPN技术等技术实现了企业网络的互通互联。在此基础上通过编程实现了基于蚁群算法的企业网络故障定位系统,并通过系统的测试分析和故障案例分析,验证了在大型企业中引用蚁群算法思想定位网络故障优于其他普通算法。
[Abstract]:This thesis is based on the research of a new computing model of congestion control for the project funded by the Natural Science Foundation of Hubei Province (No. 2009Chb008- Stream Control Transmission Protocol), whose object is the physical equipment necessary for network communication. The traditional network management is to monitor the fault phenomenon of the management object, or to receive the fault report from the management object, create and maintain the fault log record base, and analyze the fault log manually to realize the network fault location. However, there are many uncertainties in the network failure caused by various kinds of service applications, which make the artificial location of network fault bring more serious challenges. Ant colony algorithm (ACA) is one of many uncertain techniques, which is similar to the TSPV VRP model and involves network search problems. It has good positive feedback and fault tolerance, and has a good balance between complexity and accuracy. It is suitable for large-scale network environment, so this paper takes the precision, convergence and efficiency of ant colony algorithm as the target, and deeply studies and analyzes the network fault location technology of large enterprises. Firstly, this paper summarizes the commonly used network fault location techniques, analyzes the advantages and disadvantages of deterministic and uncertain technologies, and proposes a network fault location method based on ant colony algorithm. Network analyzer and network probe are introduced into large enterprise network environment, and their communication protocols are improved. In order to adapt to the dynamic nature of the fault, the time slice information is introduced, and the pheromone concentration is introduced to solve the upstream and downstream problem of the network fault. In addition, the fault location system model of large enterprise network is designed, and the correctness of the algorithm is verified by the experimental simulation under the condition that there are more network nodes, the least number of network probes are deployed and the number of fault search is minimum. Finally, taking the network environment of the large enterprise (Hubei Radio and Television Huofeng Advertising Media Library (standby) system) as an example, the paper makes a comprehensive analysis and design of the security requirements of the enterprise network. The internal network of the enterprise is built through the simulation platform GNS3, and the basic network technologies such as VLAN technology, link aggregation technology, firewall technology, are used to build the network, such as the VLAN technology, the link aggregation technology, the firewall technology, the VLAN technology, the link aggregation technology and the firewall technology. Redundancy technology VPN technology and other technologies to achieve the interconnection of enterprise networks. On this basis, the enterprise network fault location system based on ant colony algorithm is realized by programming, and the test and analysis of the system and the fault case analysis are carried out. It is verified that the ant colony algorithm is superior to other common algorithms in network fault location.
【学位授予单位】：武汉轻工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP393.07

【参考文献】