MySQL database design specification

I. Paradigm

1, 5 paradigms

• First Normal Form: atomic and cannot be split;
• Second normal form: Each data record in the data table can be uniquely identified, and non-primary key fields completely depend on the primary key.A table is an object for its own independent meaning.
• Third normal form: other non-primary key fields in the table cannot depend on other non-primary key fields, that is, non-dependent transitivity;
• Bath-Kord normal form (third normal form after optimization): the third normal form is reached, and there is only one candidate key, or each candidate key is a single attribute;
• Fourth Normal Form: Eliminate non-trivial and functionally dependent multi-value dependencies, that is, eliminate many-to-many relationships in the same table;
• Fifth Normal Form (5NF, perfect normal form): If every join dependency in a relational schema R is implied by a candidate key of R.

2. The advantages and disadvantages of the paradigm:

• Benefit: Eliminate data redundancy
• Disadvantages: reduce query efficiency, multi-table association, may be the index strategy failure

3. Denormalization

Some data may seem redundant, but they are suitable for business and need to be prioritized. At this time, de-normalization is required.

There may be a conflict between normalization and performance. You can add redundant fields to the table to reduce search time, or insert calculated columns to facilitate query.

Second, keys and related attributes

• Super key: a set of attributes that can uniquely identify a tuple;
• Candidate key: If the super key does not include redundant attributes, then this super key is the candidate key;
• Primary key: the user can choose one of the candidate keys;
• Foreign key: an attribute in one table is not the primary key of this table but the primary key of another table
• Main attribute: the main attribute involved in the candidate key
• Non-primary attribute: relative to primary attribute

Three, ER model

1. Three elements

• Entity: A data object, which can exist independently, usually a data table, represented by a rectangle;
• Attribute: the characteristics of the entity, usually a field, represented by an ellipse;
• Relation: The connection between entities, represented by a diamond.

2. Types of relationships

• One ​​to One
• One-to-many
• Many-to-many

3. Convert the ER model diagram into a data table

• An entity is usually converted into a data table;
• A many-to-many relationship is usually also converted into a data table;
• A one-to-one relationship, or a one-to-many relationship, is usually expressed by converting it into a table's foreign key, rather than designing a new data table;
• attributes are converted into table fields.