The Design And Implementation Of Server Deployment Recommendation System Based On Rule Engine

As the numbers of modules of Huatai Securities’ IT work platform continues to increase,in order to reduce the cost of data center construction and achieve high availability of IT resources,the company’s operation and maintenance department has basically completed the construction of the "Virtualized Digital Center".With the fierce growth of the servers,when using the traditional backend-based virtualization server management method,the administrator’s work intensity will increase accordingly.Therefore,we propose a server deployment recommendation platform for the operation and maintenance personnel of Huatai Securities Company.When users put the company’s server on and off the shelf,maintaining the server information in the database and using the system’s recommendation function can get the corresponding recommendation results.The system greatly improves the efficiency of daily server maintenance and ensures the success rate of server deployment.When the business of Huatai Securities’ IT work platform entered its maturity period,the rule requirements gradually occupied a large part of the business requirements.Under such circumstances,this thesis designs and develops a server deployment recommendation system based on a rule engine.Focusing on the construction of the system,we first investigated the research status at home and abroad,which provided a reference for the construct of the system.After clarifying the functional and non-functional requirements of the system,we divide the system into four modules:information maintenance module,IP recommended module,shelf recommended module,and server recommended module.For each functional module,we explain the design ideas of each module in detail through the design of class diagrams,timing diagrams,etc.And then we describe the specific business and functional design of each functional module.Then we instruct important classes and functions by showing its code.When user want to use certain recommendation function of the system,they can view a series of rules related to the function,and configure the parameters and the priority in the rules.After the configuration is complete,the rules will take effect immediately.After the user imports the data need to be calculated,he can use the recommendation function to recommend the required results.This system uses a mainstream framework for implementation,adopting the design idea of separating frontend and backend,which choose Spring Boot as the backend framework,My SQL as the database,Nginx as reverse proxy,Vue.js as the front framework,and Redis as the query cache.In addition,it use the My Batis persistence layer framework to achieve operation and access to the database.This system has developed the Easy-rules rule engine for secondary development,and added hot-deployment function to it.The rules can be configured and take effects immediately to satisfy the requirements of different businesses.