Abstract:StateGrid Jiangsu Electric Power Co., LTD., here inafter referred to as Jiangsu company, carries out cloud computing platform construction and application, promotes special work of“adding one thousand of servers to cloud platform and immigrating applications to cloud”and provides reliable cloud computing services for the construction of ubiquitous power Internet of Things. To improve the operation and maintenance capability of cloud platform,technical proposalof metrics collection and storage, metrics analysis and visualization and realtime alert are put forward, a systematic monitoring system of physical machine, platform, storage and application is built and the cloud platform operation monitoring and abnormal event realtime warning are realized.