The normal form of the database (first normal form, second normal form, third normal form, BCNF normal form) "recommended collection"

Hello everyone, meet again, I'm your friend Quanstack Jun.

Before we understand the paradigm, let's understand DatabaseThe concept of code in the library

1.Code

1.1 Oversize

It can uniquely identify an attribute or attribute group of a tuple. Any superset containing a supercode is also a supercode. Here, the unique identification tuple can be simply understood as querying the value of a certain field or several fields.A specific row of data

1.2 Candidate code

The smallest code selected from the supercode, that is, any proper subset of it cannot satisfy the condition.That is, attributes cannot be deleted.

1.3 Master Code

Select one of the candidate codes as the primary code.

2. Normal Form (NF)

Paradigm: A collection of relational schemas that conform to a certain level, in short, as far as the standard level of database table design is concerned, the paradigms are 1NF, 2NF, 3NF, BCNF, 4NF, etc., usually highHigh-level paradigms contain low-level paradigms.The design of the database is generally BCNF, and sometimes it is necessary for performance.

2.1 1 Normal Form (1NF)

1 Normal Form: The attributes of the tables in the relationship are not separable. For a simple example, there is a student table with the following fields, sid, sname, address, sid is the primary key, sname is the student's name, address is the address, and the address data may include provinces, cities, streets, etc.of.In the future, we may need to count the number of students in a city separately, and combine so much information into one attribute, that is, it does not conform to the first normal form, and the attribute should be divided into multiple fields such as province and city to meet business requirements.

2.2 2 Normal Form (2NF)

2 Normal Form: Eliminate partial functional dependencies of non-primary attributes on codes.Functional dependence: Simply put, if for each x attribute or attribute group there is a corresponding exact y value corresponding to it, the Y function is said to be dependent on x.Write x->y complete functional dependence: there is x->y, but there is no x'->y for any proper subset of x, which is called complete functional dependence.For example, there is a grade sheet with several fields, student ID, course, grade.(student number, course) -> grade, any (student number)!-> grades, (course)!-" grade, we call grade completely dependent on (student number, course) partial functional dependency: the Y function does not fully depend on x is called partial functional dependency, for example (student number, course) -> course name), the student ID has nothing to do with the course name, i.e. (course) -> course name.

2.3 3 Normal Form (3NF)

3 normal form: Eliminate the transfer function dependence of non-primary attribute pairs of codesTransfer function dependence: A relation R(U), X, Y, Z are subsets on the attribute set U, which existsX→Y and Y→Z, but Y does not determine X, that is, Y!->X and Y does not contain Z, then there is X→Z, that is, the X transfer function determines Z, and the Z transfer function depends on X.For example: commodity table: commodity name -> warehouse, warehouse!-> Commodities, warehouses -> warehouse administrators, warehouse administrators !-> warehouses (assuming a warehouse can have multiple administrators).

2.4 BCNF Normal Form (BCNF)

BCNF paradigm: Eliminate indirect functional dependencies and transfer functional dependencies between primary attributes.

3.Summary

Generally, our database design can be in 3 normal form or BCNF normal form, but in practical projects, there is always a trade-off between performance and scalability.The higher the database design standard, the better the scalability and the lower the coupling, but the performance also suffers.Therefore, sometimes it will come to 2 paradigms. In order to reduce the association between tables and speed up the query speed, various advantages and disadvantages need to be weighed by themselves.

Publisher: Full-stack programmer, please indicate the source: https://javaforall.cn/128339.htmlOriginal link: https://javaforall.cn