Design And Implementation Of PaaS Platform Back-end Management System

In the Big Data Era, in which data is the raw material, technology is the tool, and APP who can produce values is the key. To make the best use of the value of big data, it is necessary to reduce the cost of building large data applications, and avoid the waste of time and human energy in the trial-and-error stage. The external nature of big data makes the application which involves multiple business areas more complex and diverse.The multi intelligent collaboration is necessary for the value maximization of the data.While data silos and data fragmentation will hinder the full play of the value of big data,and only the cross correlation of data can fully help to understand the customers and the market,in order to generate greater value.This thesis is based on the participation in the project during my internship experience. This project is focused on the field of big data PaaS cloud platform, and based on the application of Docker container technology, to provide cloud big data application constructions, delivery and operation service for all kinds of developers. It also provides the application life cycle, the standardization of continuous integration,image building, continuous delivery and automatic operation and maintenance services The author is mainly responsible for the development of the back-end module of the project, but from the initial demand analysis of the project to the test before the project has the author's participation.The author's work includes:(1) Requirement analysis, Through consulting partners and colleagues initially identified the core needs of the system, Including the services and applications which are provided by the platform, functional requirements such as application, user management, resource management, as well as non functional requirements such as load balancing and cross platform and all of the requirements of the project have been documented.(2) Outline design, according to the result of requirement analysis, The modules,such as code construction, service deployment, user management and resource management are divided, and the logical model of the system has been established, The implementation method of each module is also preliminarily established which lays the foundation for the detailed design later.(3 )The detailed design and implementation, combining the results of the previous stage, each module is designed in detail, after that, the code implementation and technical implementation of each module are carried out to ensure the function and effect of each module achieve the target.cross compiling is used to realize cross platform use of the system.(4) System testing and validation,Through the core function and load balancing test of the system, the system can be verified that whether it can meet the requirements of the initial stage of the project.The project has now already been deployed on line. For the platform provider,they can maximize the value of the data resources and the big data ability of the platform, to ensure the safety and integrity of the data; For application developers, they can quickly build big data applications, to easily migrate across the cloud, so as to reducing marginal costs; For big data providers, they can focus more on the ability to provide and ensure quality of service,and avoid the low-level simple duplication of labor. To meet the project objectives, it's necessary to enrich the platform services and products in the later days.