Design And Implementation Of Stock Market Quotations Analysis System Based On Dubbo Middleware

In recent years,Internet finance has developed rapidly.Major Internet financial companies and securities companies are competing to develop their own stock trading systems,and the essential function of stock trading related systems is the stock market display function.Therefore,it is necessary to design and implement a fully functional and comprehensive stock market analysis system that can support multiple securities companies to help investors obtain market data.The stock market analysis system based on Dubbo middleware includes four functional modules: stock must read,stock market overview,winners list,fear and greed index.In terms of software architecture,Dubbo middleware is used to realize microservices and decouple each module.Each module is developed using spring boot framework.Dubbo middleware is used for remote interface call and load balancing between microservices,which improves the scalability and later maintainability of the system,The operation of each functional module can be monitored through the visual monitoring platform of Dubbo middleware.The front-end uses HTML5 as the development language to apply a set of interface to apps and applets to reduce the development cost.The data storage uses the combination of Redis and My SQL to realize the separation of cold and hot data.My SQL is used to store cold data persistently.Redis stores hot data,which greatly improves the response speed of the interface and reduces the storage pressure of My SQL database.Kafka message queue is used to handle the large-scale data transmission between two services to realize the decoupling of different services and ensure the success of data transmission.Nginx reverse proxy technology is used to disperse the front-end requests to different servers for processing,and the system is deployed in the form of cluster to improve the response speed of the system when accessing with high concurrency.After the system is deployed,the average response time of each functional module of the system is about 600 milliseconds,which can cope with the access pressure of 10000 TPS in the peak period.Through the implementation of cluster deployment and load balancing,the average response time of the system is reduced by 36%,and the hardware resource consumption under the access pressure of 10000 tps is about 63%.After the system was launched,it served Jingdong finance,Jingdong stock,Tianfeng securities and other Internet Finance and securities companies.The system performed well,provided stable stock market data and achieved the expected goal.