Normalization can be defined as the process of eliminating the redundancy present in the data from relational tables by decomposing or splitting a relational table into small size tables with the help of projection. The ultimate aim to do this is to have only primary keys on the left-hand side of a functional dependency.
Benefits Of Using Normalization :
- Normalization can avoid repetitive entries.
- It helps to reduce required storage space.
- Normalization can prevent the need to restructure existing tables to accommodate new data.
- It increased speed and flexibility of queries, sorts and summaries.
Types Of Normalization :
- First Normal Form (1NF)
- Second Normal Form (2NF)
- Third Normal Form (3NF)
- Boyce-Codd Normal Form (BCNF)
First Normal Form (1NF):
By definition every relational table is in first normal form. All values present in the columns are atomic.
That mean’s they contain no repeating values. First normal form is minimal requirement.
- 1NF eliminating repeating groups in each table.
- It creates a separate table for each set of related data in a table.
- It Identifies each set of related data with a primary key.
Second Normal Form (2NF):
According to the definition of the Second Normal Form, tables with composite primary keys can be present in 1NF but not in 2NF. A table to be in 2NF it should be present in 1NF.
- In 2NF there should not have any partial dependency.
- Every non-key column must be dependent upon the entire primary key.
Third Normal Form (3NF) :
The third normal form can be described as “Each an every column in a relational table is dependent only upon the primary key.
” A relational table is a present third normal form if it’s in 2NF and also every Non-key column is non-transitively dependent upon its the primary key.
- All non-key attributes should be functionally dependent only upon the primary key.
Boyce-Codd Normal Form (BCNF):
Boyce-Codd normal form is a more precise or specific version of the third normal form. BCNF deals with relational tables that are having many candidate key, overlapping candidate keys and also composite candidate keys.
- BCNF is mainly based on the concept of Determinants.
- A relational table is said to be BCNF if every determinant is candidate key.
- Redundancy causes are called update anomalies.
- Update anomalies are arise whenever information inserted, deleted and updated.