In the column-oriented database system, data is stored in columns instead of rows. The aim of this arrangement is to efficiently write the data to and read it from a hard disk. The columnar arrangement can increase the reading and writing speed and database queries can be accelerated.
In a columnar database, the column contents are directly comparable and belong to each other row by row. Individual terms can be assigned to the columns. For example, the product name, price, delivery time, etc.
One advantage offered by such a columnar representation is that the data can be compressed, for example, using mathematical functions such as the sum, maximum, minimum, average and many other functions. In addition, columnar storage requires less memory than a relational database.