基于高可靠双冗余以太网的组播及MPLS技术研究

发布时间：2018-05-06 12:19

本文选题：双冗余以太网 + 可靠性　；参考：《电子科技大学》2014年硕士论文

【摘要】：以太网传输速度高、安装容易、兼容性好,而且几乎支持所有常用的网络协议,多年来得到广泛应用,也逐渐推广到一些恶劣和极端环境下的通信任务中。在这些环境下,各硬件失效的概率增大,同时网络又要求不会因节点故障或链路失效而受影响。因此,对以太网的研究具有实际的工程应用价值。在很多工业应用场景中,网络应具有很高的可靠性以及零故障恢复时间,现在普遍通过增加网络冗余的方式来实现。然而,以太网协议和TCP/IP协议本身没有定义冗余的通信方式,因此冗余以太网通信得到广泛研究。双冗余以太网具有可行性,且可靠性高,但是在这方面的研究国内外并不是很多,很多理论和具体实现都需要进一步的研究与深入。本文依托航空科学基金课题项目“高可靠综合控制的双冗余以太网络技术研究”,以设计与实现高可靠双冗余以太网的同时进一步优化双冗余以太网、提高其性能为研究目的,开展相关的研究工作。主要研究内容和创新点如下:对相关的网络拓扑结构进行了研究、比较和分析,选择合适的拓扑结构构建网络平台,从而易于后期相关应用的实现。在高可靠双冗余以太网中应用组播,更改双冗余节点加入和离开组播组过程,同时对组播数据收发过程进行改进,从而支持不同通路模式下的组播通信。另外,在两个独立的网络中运行组播协议,形成两棵组播树,对组播数据包进行路由。通过对重要数据的冗余组播,在增加一些成本和牺牲一部分带宽的条件下,提高组播可靠性、降低其故障恢复时间。搭建网络平台并进行实验测试,验证冗余组播的可行性和可靠性。构建基于MPLS的高可靠双冗余以太网,主机节点通信之前预先建立两条路径,基于1+1保护方式,采用冗余LSP互为备份机制,把对重复数据的处理放在主机节点进行。更改主机节点的数据收发方式,使其支持冗余通信,并在不同通路模式下使用不同的LSP。在冗余通路模式下同时使用两条LSP发送相同的数据,以达到提高通信可靠性的目的。搭建基于MPLS的双冗余以太网并验证其可行性和可靠性。
[Abstract]:Ethernet has high transmission speed, easy installation, good compatibility, and supports almost all commonly used network protocols. It has been widely used over the years, and has been gradually extended to some communication tasks in harsh and extreme environments. In these environments, the probability of hardware failure increases, and the network is not affected by node failure or link failure. Therefore, the study of Ethernet has practical engineering application value. In many industrial application scenarios, the network should have high reliability and zero fault recovery time, now it is generally realized by adding network redundancy. However, Ethernet protocol and TCP/IP protocol itself do not define redundant communication mode, so redundant Ethernet communication has been widely studied. Dual redundant Ethernet is feasible and reliable, but there are not many researches in this field at home and abroad. Based on the aviation science foundation project "double redundant Ethernet Technology Research of High Reliability Integrated Control", the purpose of this paper is to design and implement high reliability double redundancy Ethernet and further optimize double redundancy Ethernet and improve its performance. Carry out related research work. The main research contents and innovations are as follows: the related network topology structure is studied, compared and analyzed, and the appropriate topology structure is selected to construct the network platform, which is easy to implement the related applications in the later stage. Multicast is applied in high reliability double redundancy Ethernet, the process of adding and leaving multicast group is changed, and the process of sending and receiving multicast data is improved to support multicast communication in different path mode. In addition, two multicast trees are formed to route multicast packets by running multicast protocols in two independent networks. Through redundant multicast of important data, the reliability of multicast is improved and the time of fault recovery is reduced under the condition of increasing some cost and sacrificing part of bandwidth. The feasibility and reliability of redundant multicast are verified by setting up the network platform and carrying out experimental tests. A high reliable double redundancy Ethernet based on MPLS is constructed. Two paths are established before the host node communicates. Based on the protection of 1 1, redundant LSP is used to backup each other, and the repeated data is processed in the host node. Change the data transmission mode of the host node to support redundant communication and use different LSPs in different path modes. In redundant path mode, two LSP are used to transmit the same data at the same time, so as to improve the reliability of communication. Build dual redundancy Ethernet based on MPLS and verify its feasibility and reliability.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.11

【参考文献】