边缘云在动力环境监控系统中的应用

发布时间：2018-05-13 01:10

本文选题：云计算 + 边缘云架构　；参考：《杭州电子科技大学》2017年硕士论文

【摘要】：动环监控系统是机房稳定运行的保障。目前云计算被运用于动环监控系统,以管理监控大面积的分布式机房。随着社会的快速发展,越来越多的厂家开始架设自己的机房,机房密度和分布面积迅速增长。而分布式云架构随服务器节点数量扩张,云服务系统消耗呈指数型上升,限制了其扩张能力。为了增强系统的扩张能力,本文提出了将边缘云架构应用于动环监控系统。边缘云架构是在分布式云架构的基础上增加了一台“核心云”用于协调各“边缘云”之间的存储、计算资源的调度,在继承了分布式云架构的基础上,解决了分布式云架构存在的一些缺陷。针对边缘云架构的特点,分析其与星型拓扑结构的契合性,提出了星型拓扑结构下的边缘云架构。通过分析其网络延迟模型,提出了一套链路选择方案为用户选择更快的连接链路,提升用户体验。针对“边缘云”分布式架构导致的协作安全维持难度大的问题,根据适用环境和边缘云架构的特点,提出了一套秘钥分配方案增加系统安全性。使用Canopy+K-means聚类对数据进行进行聚类管理。针对在Canopy算法存在的阈值T1和T2的选择问题上,分析动环监控系统的数据分布特色,提出了一套基于概率分布的计算方法进行优化。实验结果表明,本文方法能够在确保传统动环监控系统的基本功能的前提下,实现数据交互和协助计算,并能提供OMC异地操作的能力。同时在系统消耗和扩展性方面,相对于分布式云有较好的表现。
[Abstract]:The moving-ring monitoring system is the guarantee of the stable operation of the computer room. Cloud computing is currently used in mobile-ring monitoring systems to manage and monitor large-area distributed computer rooms. With the rapid development of society, more and more manufacturers begin to set up their own computer rooms. However, with the number of server nodes expanding, the consumption of cloud service system increases exponentially, which limits the expansion ability of distributed cloud architecture. In order to enhance the expansibility of the system, the edge cloud architecture is applied to the moving ring monitoring system. Edge cloud architecture is to add a "core cloud" to coordinate the storage between the "edge clouds" and calculate the scheduling of resources on the basis of the distributed cloud architecture, which inherits the distributed cloud architecture. Some defects of distributed cloud architecture are solved. According to the characteristics of edge cloud architecture, the consistency between edge cloud architecture and star topology is analyzed, and the edge cloud architecture under star topology is proposed. By analyzing the network delay model, a link selection scheme is proposed to select a faster link for the user to improve the user experience. Aiming at the difficulty of cooperative security maintenance caused by the "edge cloud" distributed architecture, a secret key allocation scheme is proposed to increase the security of the system according to the characteristics of the applicable environment and the edge cloud architecture. Canopy K-means clustering is used to manage data clustering. Aiming at the problem of selecting threshold T 1 and T 2 in Canopy algorithm, the characteristics of data distribution in moving ring monitoring system are analyzed, and a set of calculation method based on probability distribution is proposed to optimize it. The experimental results show that the proposed method can realize data exchange and assist calculation on the premise of ensuring the basic functions of the traditional moving-ring monitoring system, and can provide the ability of OMC remote operation. At the same time, the system consumption and scalability, compared with the distributed cloud has a better performance.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP277

【参考文献】