Simple and clear, the three paradigms of database design

Paradigm: Simply put, it is to standardize the design of database tables and some specifications that need to be met in order to optimize the way data is stored.
In relational databases these norms can be called paradigms.

First Normal Form

1. Each column attribute is a indivisible attribute value, ensuring the atomicity

of each column

2. The attributes of the two columns are similar or similar or same, try to merge columns with the same attributes,Make sure not to generate redundant data.
Example:

If the need is to split the address into provinces and cities,And according to its classification, then obviously the first table is not easy to meet the needs, and does not conform to the first normal form.
After splitting, example:

After splitting, it conforms to the first normal form.

Second Normal Form

Each row of data can only be associated with one of the columns, that is, a row of data does only one thing.

A table split should be split whenever duplication of data occurs in a data column.
Example:

When a person books several rooms at the same time, multiple pieces of data will be generated for one order number, so that the sub-contacts are duplicated, which will cause data redundancy.We should take him apart.
Example:

After splitting, it conforms to the second normal form.

Third Normal Form

Data cannot have transitive relationship, that is, each attribute has a direct relationship with the primary key rather than an indirect relationship.
Such a relationship between attributes like: a–>b–>c does not conform to the third normal form.

Such as Student table (student number, name, age, gender, institution, institution address, institution phone number)

In such a table structure, the above relationship exists.Student ID –> Institution --> (Institution address, institution phone number)

Such a table structure should be disassembled, as follows.

(Student ID, Name, Age, Gender, School) – (School, School Address, School Phone)