摘要
集群计算系统在大数据处理、指挥调度等领域应用越来越广泛,保障集群系统稳定运行迫在眉睫。本文针对集群内计算机管理难度大、故障影响大等问题,明确集群计算机的监控需求,研究集群计算机的监控技术,分析Cacti、Parmon、Nagios、Zenoss、Ganglia等集群监控软件的特点,通过试验对比,总结各监控软件的功能特点,探讨集群计算机监控技术的发展方向。
Since clusters have been widely applied, such as in the area of large data processing and command controlling system, it is important to maintain the stable of the clusters. Monitoring technology of the computers in cluster is researched to solve prob- lems of cluster management. The requirements of the cluster monitoring system are summarized. Monitoring technologies and the characteristics of software including Cacti, Parmon, Nagios, Zenoss, and Ganglia are analyzed. The comparison tests show that each of monitoring software has different features. The tendency of monitoring technology of computer in cluster is discussed.
出处
《计算机与现代化》
2013年第11期218-222,共5页
Computer and Modernization
基金
广东省科技计划项目(2012A080102003)
广东省省部产学研结合项目(2012B090500012)
关键词
集群
计算机
状态监控
预警
cluster
computer
state monitoring
warnings
作者简介
作者简介:吴怡风(1992-),男,四川成都人,美国伊利诺伊大学厄巴纳一香槟分校工程师,本科,研究方向:计算机应用;
归强(1981-),天津人,广东粤铁瀚阳科技有限公司工程师,硕士,研究方向:大数据集群计算,可视化;
罗明宇(1971-),四川成都人,总工程师,博士,研究方向:大数据集群计算,通信系统。