Column-oriented database management systems

In the following thesis I will present column-oriented database. Among other things, I will answer on a question why there is a need for a column-oriented database. In recent years there have been a lot of attention regarding a column-oriented database, even if the existence of a columnar database management systems dates back in the early seventies of the last century. I will compare both systems for a database management – a colum-oriented database system and a row-oriented database system and describe key features of the column-oriented database that allow faster query execution. In the practical part I will introduce two different techniques that can be used to implement a columnar-oriented database design in row-oriented database management system. This two techniques are Vertical partitioning and Index organized table (IOT). I will compare the performance of previous two techniques, row-store and a column-store and show that the row-store performance is significaly slower on analytical queries then a column-store