面向WebJoin的服务器集群监控系统的设计与实现

发布时间：2018-03-10 10:45

本文选题：服务器集群　切入点：集群监控系统　出处：《北京邮电大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着互联网技术的不断发展,尤其是移动互联网的快速兴起,使得网络接入用户数量急剧增加,这给互联网中心服务器造成了巨大的性能压力,为了应对这个问题,采用多台应用服务器组成一个服务器集群系统,从而发挥出单台应用服务器无法比拟的性能、可用性和可扩展性的集群技术开始获得广泛的应用和发展。由于服务器集群系统中服务器节点之间的独立性和松散性,对这些服务器节点进行监控、维护和异常的处理变得异常繁琐和复杂,这样服务器集群监控系统应运而生。但是随着服务器集群系统的不断演进和发展,服务器节点的数目急剧增加,其种类也越来越多样化,而且服务器集群监控的范围也从传统的服务器节点等硬件层面,拓展到操作系统、应用、服务等软件层面,监控的职责也从简单的数据采集和展示发展到集群系统异常的自动发现和自动处理。为了满足以上所提出的需求,本文设计实现了一种新的服务器集群监控解决方案。本文将服务器集群监控系统自底向上划分为三个数据层,分别是原始数据层、结构数据层和数据应用层。原始数据层主要完成监控数据的采集工作,采用异步消息的方法实现分布式数据的汇总收集；结构数据层主要完成对原始数据的初步处理,按照数据的类别整理成结构化数据,并实现异构数据的存储；数据应用层的功能是对监控数据的使用,主要完成两项功能,异常情况的发现及处理和监控数据的可视化展示。此外,本文还针对实际的服务器集群系统—-WebJoin系统,设计并实现了该服务器集群监控方案。
[Abstract]:With the continuous development of Internet technology, especially the rapid rise of mobile Internet, the number of network access users has increased dramatically, which has caused tremendous performance pressure to the Internet central server. In order to deal with this problem, Using multiple application servers to form a server cluster system, thus giving play to a single application server can not compare the performance, The availability and extensibility of cluster technology has been widely used and developed. Due to the independence and looseness of server nodes in the server cluster system, these server nodes are monitored. Maintenance and exception handling become extremely cumbersome and complex, so the server cluster monitoring system emerges as the times require. However, with the continuous evolution and development of the server cluster system, the number of server nodes increases dramatically, and the types of server cluster monitoring are more and more diversified, and the scope of server cluster monitoring is also from the traditional server nodes and other hardware level. Extended to the operating system, applications, services and other software levels, monitoring responsibilities also developed from simple data collection and display to the cluster system exception automatic detection and automatic processing. This paper designs and implements a new server cluster monitoring solution. In this paper, the server cluster monitoring system is divided from bottom to top into three data layers, namely, the original data layer, the structural data layer and the data application layer. The method of asynchronous message is used to collect the distributed data, the structure data layer mainly completes the initial processing of the original data, arranges the structured data according to the type of the data, and realizes the storage of the heterogeneous data. The function of data application layer is the use of monitoring data, mainly accomplishing two functions, the discovery and processing of abnormal cases and the visual display of monitoring data. In addition, this paper also aims at the actual server cluster system-WebJoin system. The server cluster monitoring scheme is designed and implemented.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.05;TP274.2

【参考文献】