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
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.
Professional Data Analyst and Business Intelligence Developer with experience of delivering industrial projects for Supply Chain and Insurance Industry . Sharing all my experience and insights in databases and data warehousing and open to learn from fellows !
Happy Reading and Learning!