From RDBMS to Key-Value Store: Data Modeling Techniques

In the previous blog, we found out what are the basic concepts and operations of LevelDB. In this blog, we are going to find out some techniques to model an existing RDBMS to a Key-Value schema.

Before moving on let’s have a quick recap on RDBMS and its concepts, just in case you forgot them already. 😃

1. RDBMS Recap

RDBMS or a relational database management system is based on the relational model which was introduced by E. F. Codd. In this model, the data is ordered in a structure with arranged rows and columns known as a table (relation). The rows are also referred as tuples whereas the columns are referred as attributes. The following table would be helpful in understanding this terminology.

Here employee_id, first_name, last_name and address are attributes which define a tuple (row). Following is a representation of all the tuples of this table.

(1, John, Doe, New York)
(2, Benjamin, Button, Chicago)
(3, Mycroft, Holmes, London)

The attribute employee_id is used to uniquely identify a tuple, as its value is distinct for each of the tuples. This is known as the primary key.

However, a RDBMS schema consists of several interrelated tables like this. The connection between these tables is…