基于Nagios的监控数据分析展示平台

基于Nagios已设计开发了中国科学院超级计算环境监控平台,通过监控平台可获取集群的相关运维信息,及时了解各服务器的状况,针对出现的故障问题做出快速响应。监控平台获取的数据受存储方式等因素的限制无法方便地提供给外部使用,本文针对获取的集群数据加工处理并设计开发了获取这些数据的相应接口。开发人员可方便地利用这些接口获取数据并根据自己的需求通过不同的方式进行可视化展示。基于监控接口我们开发实现了一个用于展示超级计算环境运行状况的Show平台。用户或系统管理员可通过科技网通行证登录此平台查看中科院超级计算环境的整体运维信息,同时各管理员可查看自己集群的运维数据。Show平台为用户了解环境以及各集群的运行状况提供了方便的途径。

Design and Implement a Platform Based on Monitor Data from Nagios

We designed and implemented a monitoring platform for the super computing environment of Chinese Academy of Sciences based on Nagios. Through the platform we can get operation and maintenance data of clusters, timely understand the status of each server and make a quick response on the failure problems. Data from the platform cannot be conveniently provided for the use of the external because of some factors such as the storage. The paper processed data from the monitoring platform and designed related interfaces to access the data. Developers can easily use these interfaces to get data and display it according to their own needs in different ways. Based on these interfaces we developed a platform to show operation condition of the super computing environment. Users or system administrators can use passport of the science and technology network to login the platform and check the whole operational information of the super computing environment of Chinese academy of sciences. At the same time, administrators can view operational data of their own cluster. Show platform provides a convenient way for users to understand the environment and the running status of the cluster.