Relational Databases & RDBMS

The last lesson said a database is organised. It did not say how. There is more than one answer, but one has won so completely that the whole of your syllabus is about it: put the data in tables.

1A database made of tables

A relational database is a database in which the data is organised in the form of tables.

That is the entire idea. A table is the grid you already know from a register or a spreadsheet: columns across the top saying what each piece of data is, and one row per thing you are storing.

students — one row per student

id	name	grade	percent	dob	city
1099	Veena	10	90.5	2010-11-20	Singtam
2035	Ravi	12	83.75	2008-08-13	Gangtok
3033	Tom	8	64.25	2013-02-25	Kolkata

Read across a row and you get one whole student. Read down a column and you get the same fact about everybody. Both directions are useful, and that is why the shape works.

Why the name “relational”?

Not because tables are related to each other — that comes later, with foreign keys. A table is a relation: it is the formal, mathematical name for this grid of rows and columns. So table and relation mean the same thing, and an exam may use either word.

2RDBMS — the software for them

Last lesson: a DBMS is software that creates, manages and stores a database. Narrow that to the relational kind and you get the word your syllabus uses:

An RDBMS is a DBMS application that helps us create, manage and store relational databases in digital form.

Relational Database Management System

These are the ones worth knowing by name:

MySQL

Free and open source. The one this course uses.

Oracle

Large commercial systems — banks, airlines.

MariaDB

A community fork of MySQL, very similar to use.

MS Access

Part of Microsoft Office, for small databases.

OpenOffice Base

The free equivalent of MS Access.

SQLite

Tiny. It is inside your phone right now.

DBMS vs RDBMS

	DBMS	RDBMS
How data is stored	In files, or any structure the software chooses	In tables — rows and columns
Relationships between data	Not necessarily supported	Supported, using keys
Keys and constraints	Usually not available	Primary key, unique, not null and the rest
Typical size	Small amounts of data	Small to very large
Examples	A plain file store, early flat-file systems	MySQL, Oracle, SQLite

Every RDBMS is a DBMS

The R only adds a restriction: it stores its data as tables. So every RDBMS is a DBMS, and not every DBMS is an RDBMS. If a question asks you to name a DBMS, MySQL is a perfectly correct answer.