Development And Application Of Operation Management System For Cloud Service Platform

With the development of the Internet and the popularity of cloud computing,cloud service platforms have gradually become mainstream replacements for IT infrastructure and applications.And cloud service platform,with its flexible interaction mode,on-demand services and ubiquitous network,has subverted the business model of traditional IT services.The cloud service platform has a large infrastructure which involves many application servers,distributed databases,and network devices.So traditional manual operation and maintenance methods are inefficient and unreliable in this case,and it is difficult to guarantee the quality of service.The platform has put forward higher requirements for operation technical support to ensure the operation efficiency and application service quality of the automotive and parts industry cloud service platform developed by the team.According to the operations of the cloud service platform for automobile and parts industry,the thesis collates the business and specific functional requirements of the operation management system based on the cloud service platform for automobile and parts industry.The thesis studies the key technologies involved in the operation management system,and compares them.The thesis designs a set of operation solutions covering the functions of monitoring and alarm,automatic configuration,integrated deployment.The operation management system based on Django Web framework is implemented by layered architecture design,which includes six modules: continuous integration management,task management,monitoring management,resource management,log management and system management.For the statistical analysis of various monitoring data in the system,the ECharts open source framework combined with Ajax technology and the JSON data interaction format is used to realize the visualization of the chart which can provide some help for decision-making of operation personnel.In order to reduce duplicate alarm,K-means clustering algorithm particle swarm optimization algorithm are selected to cluster the alarms.The operation management system is applied to the cloud service platform of the automobile and parts industry.The operation management system developed in this thesis has two characteristics:(1)deepening the relationship between operations and development to ensure quality of application system by continuous integration;(2)an unified resource allocation and integrated operation platform with integrating multi-party excellent operation tools.