Design And Implementation Of Integrated Query Middleware About Relational Database And Non-Relational Database For Structure Safety Monitoring

With the rapid development of Internet technology, we are entering the era of Big Data:the new emerging data sources, easier data replication and transmission, continuous backlog of historical datas to be stored and processed. Existing relational database and non-relational database storage solutions have their own advantages, so the solution mixing SQL database and NoSQL database emerged. However, in this storage resolution, datas in the same application program need to be separated and stored separately in the SQL database and NoSQL database according to the characteristics of the data. So it involves the management of multiple heterogeneous data sources. This thesis is under the above background to conduct research on key technologies of integrated query about SQL database and NoSQL database.In this thesis, the distributed problems of the storage architecture for structure safety monitoring are introduced firstly, explaining the significance of integrated query middleware about relational database and non-relational database. By analyzing the current situation and development trend in recent years of domestic and foreign storage architecture, the research direction is proposed. Then the resolution of integrated query about SQL database and NoSQL database is focused on and the integerated query middleware architecture about SQL database and NoSQL database combining SQL master database query resolution with independent middleware resolution is summed up. Research on the NoSQL data relationalization in the architectural requirements is conducted and an improved NoSQL data transformation resolution based on the pre-inquiry is designed, by analyzing the direct triple representation algorithm of NoSQL data. Research on the NoSQL data reconstruction in the architectural requirements is conducted and an improved NoSQL data reconstruction resolution based on the pre-screening is designed, by analyzing the defects of equivalence connection reconstruction algorithm. Then, the integrated query middleware about SQL database and NoSQL database using the new improved resolutions is implemented, allowing users to transparently access to heterogeneous data sources. Finally, a test platform is set up to test on middleware functions, NoSQL data transformation algorithms and NoSQL query mode conversion algorithm. Then the performance differences between the original algorithm and the new resolution are compared and analyzed.Experiments show that the designed integrated query middleware about SQL database and NoSQL database architecture is feasible. The developed middleware system following this architecture allows the user to build standard SQL statements to access SQL and NoSQL datas uniformly without concern for different data structures of heterogeneous data sources. After experimental analyses, the improved NoSQL data transformation resolution and NoSQL query mode conversion resolution can significantly improve query performance, which have certain pratical value and engineering significance.