Introduction
As an Engineer working with data, your role involves working with relational databases, implementing BI solutions, and supporting analytical queries.
To optimize database performance and manage data productively, understanding key concepts like normalization is essential.
I dive into it.
Database Design
Database design involves structuring a database in a way that reduces data redundancy and improves data integrity. Effective database design ensures efficient data retrieval and management, which is crucial for BI and analytics applications.
Note: Proper database design is the backbone of any data-driven application. It impacts performance, scalability, and maintainability of the data system.
Normalization
Normalization is the process of organizing data in a database to reduce redundancy and improve data integrity. It involves decomposing tables into smaller tables and defining relationships between them.
First Normal Form (1NF)
First Normal Form (1NF) ensures that the table is structured such that each column contains atomic (indivisible) values, and each record is unique.
Rules
Each table cell should contain a single value (no repeating groups or arrays).
Each record needs to be unique.
Before 1NF
| EmployeeID | Name | Phone Number |
| 1 | John Doe | 1234567890, 0987654321 |
| 2 | Jane Doe | 5555555555 |
The table above violates 1NF because the Phone Number column contains multiple values.
After 1NF
| EmployeeID | Name | Phone Number |
| 1 | John Doe | 1234567890 |
| 1 | John Doe | 0987654321 |
| 2 | Jane Doe | 5555555555 |
Application
Ensuring 1NF is crucial in designing databases where data retrieval and updates need to be efficient and straightforward. By eliminating repeating groups, databases can avoid the complexities from dealing with arrays or lists within columns.
Second Normal Form (2NF)
Second Normal Form (2NF) builds on 1NF by ensuring that all non-key attributes are fully functionally dependent on the primary key. It eliminates partial dependencies, where an attribute depends only on a part of a composite primary key.
Rules
The table must be in 1NF.
All non-key attributes must be fully functionally dependent on the primary key.
Before 2NF
| OrderID | ProductID | ProductName | Quantity |
| 1 | 1 | Laptop | 10 |
| 1 | 2 | Mouse | 25 |
In this table, both (OrderID, ProductID) form the composite primary key. However, ProductName is partially dependent on the ProductID attribute alone.
After 2NF
Orders Table
| OrderID | ProductID | Quantity |
| 1 | 1 | 10 |
| 1 | 2 | 25 |
Products Table
| ProductID | ProductName |
| 1 | Laptop |
| 2 | Mouse |
Application
2NF is particularly useful in systems where relationships between different entities (like orders and products) need to be clearly defined. It helps in maintaining data integrity and avoids anomalies during insertions, updates, or deletions.
Third Normal Form (3NF)
Third Normal Form (3NF) further refines 2NF by ensuring that all the attributes are only dependent on the primary key and not on other non-key attributes, eliminating transitive dependencies.
Rules
The table must be in 2NF.
All non-key attributes must be directly dependent on the primary key.
Before 3NF
| EmployeeID | Name | Department | DepartmentLocation |
| 1 | John Doe | IT | Building A |
| 2 | Jane Doe | HR | Building B |
The table above has a transitive dependency: DepartmentLocation depends on Department, which in turn depends on EmployeeID.
After 3NF
Employees Table
| EmployeeID | Name | DepartmentID |
| 1 | John Doe | 1 |
| 2 | Jane Doe | 2 |
Departments Table
| DepartmentID | Department | Location |
| 1 | IT | Building A |
| 2 | HR | Building B |
Application
3NF is critical for complex systems where detailed relationships between entities need to be captured without introducing redundancy. It ensures that changes in non-key attributes are confined to a single table, simplifying maintenance.
Boyce-Codd Normal Form (BCNF)
Boyce-Codd Normal Form (BCNF) is a stricter version of 3NF. A table is in BCNF if it is in 3NF and every determinant is a candidate key. BCNF addresses certain anomalies that 3NF does not address.
Rules
The table must be in 3NF.
For any functional dependency X → Y, X should be a super key.
Before BCNF
| StudentID | Course | Instructor |
| 1 | Math | Mr. Smith |
| 2 | Science | Mr. Johnson |
The table above has a dependency where Course determines Instructor, which is not good.
After BCNF
Students Table
| StudentID | CourseID |
| 1 | 1 |
| 2 | 2 |
Courses Table
| CourseID | Course | InstructorID |
| 1 | Math | 1 |
| 2 | Science | 2 |
Instructors Table
| InstructorID | Instructor |
| 1 | Mr. Smith |
| 2 | Mr. Johnson |
Application
BCNF is especially useful in large enterprise systems where complex relationships and dependencies need to be managed properly. It ensures the highest level of data integrity and consistency by addressing even the smallest anomalies.
Conclusion
Normalization is a very important process in database design, ensuring data integrity and reducing redundancy. By adhering to normalization principles up to BCNF, you can design databases that are efficient, scalable, and easy to maintain.