First you have to understand what you mean by data.

In this context, data may refer to the following example: Sales data of a multinational company
XYZ. XYZ has say about 10000 different stores in 9 different countries. Each store can be classified as one of the 5 types depending on the size and one of the 3 types depending on the profit generated.

The CEO of XYZ wants to find which stores to two given countries are of TYPE 1 and provides income of type 3 and are closed on every Friday; so that he can match with some store in another company. I hope you understand the complexity of the situation. DBMS is the whole system of maintaining the above described data securely and reliably (into a disk storage) so that it can be retrieved later for any kind of comparison that may be though of.

The theoretical part of this system (RDBMS, Data Modelling) consists of drawing tables on a piece of paper and establishing different relations between tables. The practical part of this is done through a language called SQL. Through SQL (and several underlying scripts) we store data into the disk and later retrieve/ update/ delete them. This might be said as the rough definition of DBMS. But technology keeps changing every second. The whole concept is too vast to be explained in just one post. But I hope this makes sense.