SQL and NoSQL Object-Database Mapping Using Property Graphs in Relational Cases

This paper discusses the inherent challenges of SQL and NoSQL object mapping, primarily focusing on the heterogeneity of data models and its impact on database relationship. In NoSQL databases, relationships between entities can be defined using references or embedded objects. Each method has its advantages and trade-offs, and different NoSQL DBMS adopt different strategies based on their design principles and use cases. We examine how different SQL Database Management Systems (DBMS) handle relationships between entities, specifically comparing the use of embedded objects in MongoDB with the graph-based reference system in Neo4J. Object mapping is built using property graphs that represent the schema in the database. The framework was tested using the hospital information system medical record module. The results of this study show that the object mapping that has been built can run query commands on three distinct types of databases: MySQL, MongoDB, and Neo4j.