Data Modelling is the process of organising data elements into entities and mapping them into real world objects . This defines how different entities will behave together and what will be the relationship among them. Database Management Systems are designed to store information and data for softwares . Before , implementation of the system the DBMS is logically structured and designed. Because, this decides how data will be stored , processed and interact.
Previously ,we studied about proposed data models and last of it was relational database which is being mostly used in industry now . Business analyst collects requirements from client with careful and elaborate discussions. Therefore , This leads to data modelling diagrams and decisions about the business logic .
Data Modelling comes in pretty much handy in implementing those decisions . Think of data modelling as a road map into your software solution. Because , that’s the role of it in DBMS implementation.
Concluding above discussion and a simple definition for Data Modelling could be :
The logical structure of a database and determining the manner in which is stored, organised and manipulated
Two Data Models are
- Entity-Relationship Model ( ERD Diagrams )
- Relational Model
Entity & Members
Entities ( real-world ) objects and their attributes (properties ) . Example Student is an entity and his Name , age , address , gender etc are attributes
The logical association and way of interaction between entities enforced by using cardinalities and modalities . Example , Student studies in a University . When we link them we need to define how should they interact. Because , data modelling requires that a relationship between entities must have a minimum & maximum association. For instance , here we need to know how many students can be in a university at maximum and minimum.
This is all linked together by Crow’s Foot Notation . Check a saple ERD for a University Management System
RELATIONAL MODEL :
This model is more scientific . Hence , its implementation is very wide and practical in the industry . Here , entities are known as relation/table . Attributes/members are converted into columns. Every members has its own domain data in it. See below for the explanation :
Table : Relation also referred as entity in ERD Model
Attributes : Properties of entities
Columns : For storing attribute data
Rows : For unique data against all attributes
These relations in a DBMS are usually combine together by Joins , Normalised and constraints for reporting , storage and support of application.
Also See : Joins Explained
An example for Relational Model given below. We will use same example which we were using for an ERD .
Here we define attribute name with the data type and relationship between them with cardinality and modality.
In the above , link we have a case study. We will first breakup the scenario into elements & make an ERD out of it. Classic example of data modelling .
Once you are done with these articles. You will be able to translate the complete user requirement into database management system because data modelling is an essential to all other concepts.
If this was helpful , follow us or subscribe and share with your friends.