Deciphering Data Architectures

Chapter 8. Approaches to Data Modeling

Data modeling is a high-level conceptual technique used to design a database. There are several different types of data modeling, including relational modeling and dimensional modeling. The process involves identifying the data that needs to be stored, then creating a structured representation of that data and the relationships among the data, organized into tables and columns. Think of the tables and columns as the logical representation of the database, where the physical data stored in those tables and columns can be in a relational database product or a data lake.

You can use data modeling with any type of database: relational, dimensional, NoSQL, and so on. I recommend dedicating a lot of time to data modeling to ensure that your database is logical, efficient, and easy to use. This ensures maximum performance and makes it easier to retrieve and analyze data.

Relational Modeling

Relational modeling, developed by Edgar F. Codd in 1970 (as mentioned in Chapter 2), is a detailed modeling technique used to design a database. It involves organizing the data into tables and defining the relationships between the tables. In relational databases and relational data warehouses, each table consists of rows (also called records or tuples) and columns (also called fields or attributes). Each row represents a unique instance of the data, and each column represents a specific piece of information about the data.

Keys

In relational modeling, you ...