Reflection on MongoDB Database Logical and Physical Modeling

Traditional relational database design uses conceptual, logical, and physical modeling based on Peter Chen's methods. UML (Unified Modeling Language), though being a set of software development tools, is often used to conceptually model data and relationships. This paper presents a methodical approach to logically and physically model data in MongoDB by utilizing UML conceptual modeling aids. Application of data models greatly impacts performance, scalability and flexibility of data systems. Furthermore, application life-cycle and expansion impact long-term decisions made during modeling. This paper takes into consideration application requirements, access patterns and life-cycle during logical and physical modeling in a cloud environment. The flexibility that a NoSQL modeling paradigm embodies makes logical and physical modeling complex, with no hard-and-fast rules. This paper presents logical and physical modeling concepts required during designing database applications with MongoDB.