Research On Access Control And Upload Optimization Of Enterprise Storage System Based On Microservice

Enterprise data continues to grow with the development of the business and the expansion of the scale.The centralized storage of data can solve the problem of information islands within the enterprise,facilitating the collaborative office and mobile office.In some internal collaboration systems,users of different systems need to access enterprise data and operate shared data after authentication and authorization.Enterprise applications such as learning management systems and game management systems need to deliver unstructured file storage functions to reliable and stable enterprise storage systems.Therefore,enterprise storage systems need to consider two key issues: strict data security access control and stable and reliable access performance.Based on the analysis of the problems encountered in the practical use of enterprise storage system developed by Communication & Computer Network Key Laboratory of Guangdong,this paper investigates the solutions of other companies at home and abroad,and proposes to provide better local expansion ability and access performance through the microservice split system.Firstly,this paper determines the service boundary according to business rules,and carries out microservice migration practice according to chimney principle and incremental migration principle.Secondly,according to the requirement of storage function provided by the enterprise storage system for other applications,based on OAuth2.0 and JWT,this paper designs a platform-based security access model and access control strategy of system interface,which manages the access rights of open interface and ensures the security and controllability of system access.Finally,this paper innovatively proposes an uploading performance optimization method based on the memory file system.Through the server running memory monitoring and uploading service memory usage settings,a dual-guaranteed dynamic routing adaptive upload processing strategy is designed and implemented.Experiments show that this method can effectively reduce the response time,improve the throughput rate,and reduce the error rate in the case of high concurrent upload requests compared with ordinary upload schemes.The enterprise storage system implemented in this paper has been tested and run in Communication & Computer Network Key Laboratory of Guangdong for half a year.Both iOS client and Windows client of the application store have supported the version system of microservice.The microservice migration model and key problem-solving methods proposed in this paper provide practical modes and solutions for microservice practice of other enterprises.The user authentication method of the old system microservice migration and the upload optimization method based on the memory file system have been applied for two patents.In addition,this paper also makes a reasonable vision and prospect for the research direction of service running load monitoring and automatic capacity expansion,and providing personalized uploading services based on information such as uploading behavior.