网络虚拟化管理平台故障恢复模块的设计和实现

发布时间：2018-10-17 10:15

【摘要】：随着互联网技术的蓬勃发展,互联网已成为人们生活和工作中重要的基础设施。为了解决当前网络架构的僵化问题,网络虚拟化技术受到了学术界广泛关注。为了更好的管理和运行网络虚拟化环境,需要一个网络虚拟化管理平台。网络虚拟化管理平台集成了网络虚拟化环境下的许多管理和调度功能,方便网络运营者和用户管理和使用网络虚拟化环境。为了保证网络虚拟化环境下应用和业务运行的稳定性,需要在网络虚拟化管理平台中设计和实现一个故障恢复模块。本文工作主要是设计和实现网络虚拟化管理平台故障恢复模块,故障恢复模块作为网络虚拟化管理平台中核心功能模块之一,主要解决底层物理链路发生故障后引发的虚拟网络故障恢复问题。本文首先研究了两种故障恢复策略,即主动故障恢复策略和被动故障恢复策略,主动故障恢复策略会为每条虚拟链路计算一条备份路径,如果某条链路发生故障,只需将该链路迁移到备份路径上即可,如果该链路没有合适的备份路径,则调用被动故障恢复策略;此外,通过分析网络虚拟化环境下的资源类型以及属性,构建了网络资源模型,并且该模型支持故障恢复策略的执行;根据故障恢复的流程,本文将故障恢复模块划分为六个功能点,分别是虚拟请求定制、基本调度、链路备份、故障监听、故障告警以及故障恢复;接着,本文在故障恢复策略和资源模型的基础上,依据软件工程方法,依次对故障恢复模块的功能点进行需求分析、概要设计、详细设计和功能实现;最后本文对故障恢复模块进行了功能测试。
[Abstract]:With the rapid development of Internet technology, Internet has become an important infrastructure in people's life and work. In order to solve the rigid problem of current network architecture, network virtualization technology has received extensive attention in academic circles. In order to better manage and run network virtualization environment, we need a network virtualization management platform. The network virtualization management platform integrates many management and scheduling functions in the network virtualization environment to facilitate network operators and users to manage and use the network virtualization environment. In order to ensure the stability of application and service operation in network virtualization environment, it is necessary to design and implement a fault recovery module in network virtualization management platform. The main work of this paper is to design and implement the fault recovery module of the network virtualization management platform, which is one of the core function modules in the network virtualization management platform. It mainly solves the problem of virtual network fault recovery caused by the failure of the underlying physical link. This paper first studies two kinds of fault recovery strategies, namely, active fault recovery strategy and passive fault recovery strategy. The active fault recovery strategy computes a backup path for each virtual link, and if a link fails, Simply migrate the link to the backup path, call the passive fault recovery strategy if the link does not have an appropriate backup path, and analyze the resource types and properties in the network virtualization environment, The network resource model is constructed, and the model supports the execution of fault recovery strategy. According to the flow of fault recovery, the fault recovery module is divided into six functional points, namely virtual request customization, basic scheduling, link backup, and so on. Then, based on the fault recovery strategy and resource model, according to the software engineering method, the function points of the fault recovery module are analyzed and designed in sequence. Detailed design and functional implementation; at last, the function of the fault recovery module is tested.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP393.07

【相似文献】