Column-oriented database

What is columnar database?
A columnar database is a database management system (DBMS) that stores data in columns instead of rows like relational DBMS do. The main differences between a columnar database and a traditional row-oriented database focus on performance, memory requirements, and schema modification methods. The goal of this type of database is to efficiently read data from secondary storage and write it to secondary storage to speed processing time when returning a query.

A columnar database can also be referred to as a columnar database

The data that is saved appear in record order, ie the entry in the first column relates to the entry in the second column and other columns if the entry appears in the same row. For example, the 100th entry in column 1 belongs to the same entry as the 100th entry in column 2. This makes it possible to access individual data in columns as a group and not access them individually and line by line. Data is stored top-to-bottom rather than left-to-right as in a row-oriented RDBMS.

Columnar databases are advantageous because the data can be highly compressed, allowing column operations to be performed at high speed. It is also self-indexing and takes up less disk space compared to an RDBMS. However, the loading process may take some time, depending on the size of the data. It also becomes a challenge to the database when incremental loads are encountered, which leads to system performance issues.