Design And Implementation Of Metadata-driven Reactive Online Education System

It's one of the core businesses for financial institutes such as CFFEX(China Financial Futures Exchange)to do investment knowledge training to investors,for example the finantial futures knowledge training.But as for now,this kind of business still stays off-line for most of them.As smart phones in everyone's hands,it's quite inefficient since the off-line courses are unable to replay and only accept a few users each time.On the other hand,as the mobile era comes,some new demands come such as a booming page hints,multiple user devices and better user experience.Hence it's the demanding business needs for those financial exchanges to build mobile based online training web sites to accept multi-level investors as their guests.To meet the needs in building online education system of CFFEX,this paper did a deeper investigation on reactive technologies from its features and some related frameworks.And then it came up with the design solutions on how to build a reactive online education platform.In the solution,it separates the frontend UI from the backend services.For frontent UI,it gives a metadata driven design approach on how to build frontend SPA(single page application)with Angular directives.In building the backend service,this paper utilizes the CQRS(command query responsibility segregation)design pattern to improve the concurrency ability of the system.In persistence aspect,it uses Event Store to persistent the domain events and take Mongo as the query database.The main work of this paper is as follows:1.The system requirement analysis of investors' online education system The paper first described the business scenario of the investors' online education platform,then it gave the requirement functional tree and business use cases from different business modules and at last laid out the non-functional requirement of the system.2.The design of the reactive online education platform architecture.After the analysis of system requirement of online education platform,the paper came up with the system architecture by separating the system into two parts: the frontend UI and the backend services.For the frontend UI,it utilized the metadata driven approach in implementing the responsive SPA(single page application),for the backend service,it utilized the CQRS design approach in improving the system performance.At last it gave the deployment architecture diagrams and described how the system was deployed.3.The design and implementation of the reactive online education platform.In this part,the paper discussed the details of the system design and implementation from 4 different aspects: A)Resolved the authentication problem by implementing the JWT authentication solution.B)The implementation of the reactive frontend based on the metadata driven design approach.C)Finished the reactive REST API and routing strategy design and implementation.D)Introduced reactive service implementation based on the CQRS design approach.4.The online education platform testing and result analysisIn this part,the paper first gives the introduction to testing method and testing environment configurations,after that it verified the platform from three different perspectives: the platform functional,performance and foult tolence.From functional perspective,it verified the website content management module and online course management and course learning module and verified the responsive web page design.From performance perspective,it did stress testing on Query API and command API and analysised its testing results.From foult-tolance perspective,it did a fail-over tesing and analysised its results.The testing result verified the feasibility of the design approach of the reactive online education platform.The design approach of metadata driven reactive online education platform this paper comes up with,realized the system requirement of CFFEX by providing high availability and responsive ability.Especially the use of event-driven concurrent framework with event sourced persistence methodology and CQRS design pattern,dramatically improves the concurrency efficiency and hence improves the system scalability.