网络节点并发数据分析技术研究

发布时间：2018-08-31 16:44

【摘要】：随着计算机应用技术的普及和虚拟化技术的快速发展,承载业务信息的网络已经渗透到人们生活的各个领域。网络往往由大量的功能节点(网络设备或终端)相连组成。如何有效及时地发现网络中所有节点、实时监测节点的异变状态和异常事件、动态感知网络拓扑结构、识别云体系架构下节点拓扑,就成为网络自治域系统可靠运行的主要问题。本文从传统网络节点监测技术出发,深入分析了网络节点数据异变监测技术和网络拓扑结构计算技术,明确了新一代网络节点监测系统的客观要求,包括获取复杂网络动态拓扑、节点数据异变实时监测和云体系下虚拟节点拓扑监测。研究并改进了网络节点并发数据分析技术。研究基于SNMP协议和BFD协议重点研究了网络节点并发数据分析关键技术,主要进展包括节点身份唯一标识及类型辨别、虚拟节点和隐含节点发现技术的提出、基于三表网络拓扑计算技术的提出、链路层隐含节点拓扑计算技术的提出、基于BFD协议网络拓扑增补技术的提出、报警信息采集上报机制的优化。根据实现功能和优化性能设计了系统各个模块,达到对网络所有节点运行状况实时地监测和节点异常及时地上报,对云体系架构下节点拓扑结构初步识别,帮助管理者及时感知异常并有效地解决问题。基于以上工作,围绕系统的兼容性、健壮性和高效性,在真实网络环境中测试系统功能和性能。测试结果表明,系统能够及时掌握网络节点运行状态和节点异常事件,提高了拓扑结构的真实性,对网络系统安全运行具有重大作用。
[Abstract]:With the popularization of computer application technology and the rapid development of virtualization technology, the network carrying business information has penetrated into all areas of people's lives. The network is often composed of a large number of functional nodes (network equipment or terminals) connected. How to effectively and timely discover all nodes in the network, real-time monitoring of the node's mutation status and Anomalous events, dynamic perception of network topology and identification of node topology in cloud architecture are the main problems in the reliable operation of network autonomous domain system.Based on the traditional network node monitoring technology, this paper analyzes the network node data mutation monitoring technology and network topology computing technology, and defines the new generation network node. The objective requirements of point monitoring system include acquiring dynamic topology of complex network, real-time monitoring of node data variation and virtual node topology monitoring in cloud system. Including the unique identification and type identification of nodes, virtual node and hidden node discovery technology, based on the proposed three-table network topology computing technology, link layer hidden node topology computing technology, based on the BFD protocol network topology updating technology, alarm information collection and reporting mechanism optimization. Each module of the system is designed to monitor the running status of all the nodes in the network and report the anomalies in time. The topological structure of the nodes under the cloud architecture is identified preliminarily, which helps managers to perceive the anomalies in time and solve the problems effectively. The test results show that the system can grasp the running state of network nodes and abnormal events in time, improve the authenticity of topology structure, and play an important role in the safe operation of network system.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP393.0

【相似文献】