How to distinguish relational databases from column-oriented data storage systems?

Relational databases (RDBs) store data in the form of tables with clearly defined rows and columns, where each row is a record and each column is an attribute. Data is organized by rows, which is convenient for transactions and operations with individual records.

Column-oriented storage systems (columnar DBMSs) store data by columns rather than by rows. This optimizes read and aggregation operations over large volumes of data, especially in analytical tasks where individual columns are often read.

Main differences:

• Storage structure: RDBs are row-based, columnar DBMSs are column-based.
• Optimization: RDBs are better for OLTP (transactional processing), columnar DBMSs are better for OLAP (analytics).
• Performance: Columnar systems are faster for queries on individual columns and aggregations.

Example: In a relational database, to fetch all values of a single column, all rows must be read, whereas in a columnar database, the required column is immediately accessible.