MySQL advanced learning notes (4)

One 、 Database design specifications

If it is a poor database design, it may cause the following problems :

• data redundancy 、 Repetition of information 、 Waste of storage space
• Data update 、 Insert . Deleted exception
• Can't represent information correctly
• Loss of valid information
• Poor program performance

Good database design has the following advantages :

• Save data storage space
• It can guarantee the integrity of data
• Convenient for the development of database application system

All in all , When you start setting up the database , We need to pay attention to the design of data table . In order to establish less redundancy 、 A well structured database , There are certain rules that must be followed when designing a database .

1. normal form （Normal Form）

• First normal form （1st NF）

First normal form It is mainly to ensure that The value of each field must be atomic , That is to say, in the data table The value of each field is the smallest data unit that cannot be split again .

• Second normal form （2nd NF）

The second paradigm requires , On the basis of satisfying the first paradigm , Also meet every data record in the data sheet , Are uniquely identifiable . And all non primary key fields , Must be completely dependent on the primary key , You can't rely on only part of the primary key . If you know the values of all attributes of the primary key , Any tuple can be retrieved ( That's ok ) Any value of any property of .( The primary key in the request , In fact, it can be extended and replaced with candidate keys ).

Second normal form (2NF) requirement The attribute of an entity depends entirely on the primary keyword . If there is an incomplete dependency , Then this attribute and this part of the main keyword should be separated to form a A new entity , There is a one to many relationship between the new entity and the meta entity .

• Third normal form （3rd NF）

The third paradigm is based on the second paradigm , Ensure that every non primary key field in the data table is directly related to the primary key field , That is to say, all non primary key fields in the data table cannot depend on other non primary key fields .( namely , There cannot be a non primary attribute A Dependent on non primary properties B, Non primary property B Depends on the primary key C The situation of , There is "A→B→C The decisive relationship of ) informally , This rule means that all non primary key attributes cannot have dependencies , Must be independent of each other , The candidate key can be extended here .

• BCBF（ Bass paradigm ）

stay 3NF On the basis of improvement , Put forward Bass paradigm (BCNF) , It's also called bath - The COD paradigm (Boyce-Codd Normal Form).BCNF It is considered that no new design specification has been added , It only requires stronger design specifications in the third paradigm , Make the database redundancy smaller . therefore , It is called the modified third paradigm , Or extended third paradigm ,BCNF It is not called the fourth paradigm .
If a relationship reaches the third paradigm , And it There is only one candidate key , perhaps Each of its candidate keys is a single attribute , Then the relationship naturally reaches BC normal form . It's in 3NF In addition, it eliminates the partial or transitive dependency of primary attributes on candidate keys
Generally speaking , A database design conforms to 3NF or BCNF That's all right. .

• Fourth normal form

Multivalued dependency The concept of :

• Multivalued dependency namely One to many relationships between attributes , Write it down as K→→A.
• Function dependency In fact, it is Single value dependency , Therefore, the one to many relationship between attribute values cannot be expressed .
• Ordinary multi value dependence : The complete U=K+A, One K Can correspond to multiple A, namely K→→A. At this point, the whole table is a set of one to many relationships .
• Non trivial multivalued dependence : The complete U=K+A+B, One K Can correspond to multiple A, It can also correspond to multiple B, A And B Independent to each other , namely K→→A,K→→B. The entire table has multiple sets of one to many relationships , And there are :“ One “ Part is the same set of attributes ,“ many “ Part is a collection of independent attributes .

The fourth paradigm is to satisfy bath - The COD paradigm (BCNF) On the basis of , Eliminate nontrivial and non functional dependencies with multivalued dependencies ( That is, delete the many to many relationship in the same table ).

• The fifth paradigm 、 Domain key paradigm

In addition to the fourth paradigm , We also have a more advanced fifth paradigm ( Also called perfect paradigm ) And domain key paradigm (DKNF).
In meeting the fourth paradigm (4NF) On the basis of , Eliminate connection dependencies that are not implied by candidate keys . If the relationship pattern R Each connection dependency in is determined by R Implied by the candidate key of , It is said that this relationship model conforms to the fifth paradigm .
Functional dependency is a special case of multivalued dependency , Multivalued dependency is actually a special case of connection dependency . However, unlike functional dependency and multivalued dependency, connection dependency can be derived by semantic connection , It is In relational connection operation Only then reflects . The relational pattern with connection dependency may still encounter data redundancy and insertion 、 modify 、 Delete exceptions and other problems .
The fifth paradigm deals with Lossless connection problem , This paradigm has little practical significance , Because lossless connections rarely occur , And imperceptible . The domain key paradigm attempts to define an ultimate paradigm , This paradigm considers all dependency and constraint types , But the practical value is also the smallest , Only exists in theoretical research .

• The advantages and disadvantages of paradigms

The advantages of paradigm : Standardization of data helps Eliminate data redundancy in the database , Third normal form (3NF) It's usually considered to be in performance 、 The best balance between scalability and data integrity .
The drawbacks of paradigms : The use of paradigms , Probably Reduce the efficiency of query . Because the higher the paradigm level , The more data tables you design 、 The more refined , The less redundant the data is , When querying data, you may need to associate multiple tables , It's not only expensive , It may also invalidate some indexing strategies .

Paradigms just put forward the standard of design , In fact, when designing data tables , It is not necessary to meet these standards . In development , We will encounter performance and read efficiency violations The principle of anti paradigm , Improve the read performance of the database by adding a small amount of redundant or duplicate data , Reduce associated queries ,join Times of table , Realize the purpose of space for time . Therefore, in the actual design process, we should combine theory with practice , Flexible use of .

2. Anti normalization

Anti paradigm can be achieved by Space for time , Improve the efficiency of query , But anti paradigm also brings some New problems :

• The storage space has become larger
• A field in a table has been modified , Redundant fields in another table also need to be modified synchronously , Otherwise, the data is inconsistent
• If you use stored procedures to support data updates 、 Delete and other extra operations , If updates are frequent , It will consume system resources
• In the case of a small amount of data , The anti normal form does not embody the performance advantage , It may also complicate the design of the database

When redundant information is valuable or can greatly improve query efficiency , We will adopt anti paradigm optimization .

• Suggestions for adding redundant fields

Adding redundant fields must meet the following two conditions . Only if these two conditions are met , You can consider adding redundant fields .

1. This redundant field does not need to be repaired frequently
2. This redundant field is indispensable for query .

Anti paradigm optimization is also commonly used in Data warehouse design in , because Data warehouses usually store historical data , The real-time requirement of addition, deletion and modification is not strong , Strong demand for historical data analysis . In this case, appropriate redundancy is allowed , More convenient for data analysis .
The difference between data warehouse and database in use : .

1. The purpose of database design is to capture data , The purpose of data warehouse design is to analyze data ;
2. The database has strong real-time requirements for data addition, deletion and modification , Need to store online user data , Data warehouse stores historical data ;
3. Database design needs to avoid redundancy as much as possible , However, in order to improve query efficiency, some redundancy is also allowed , and The design of data warehouse is more inclined to adopt anti paradigm design .

3.ER Model

ER The model is also called Entity relationship model , It is used to describe the objective things in real life 、 The nature of things , And a data model of the relationship between things . In the design stage of developing information system based on Database , Usually use ER Model to describe information requirements and information characteristics , Help us clarify the business logic , So as to design an excellent database .
ER There are three elements in the model , Namely Entity 、 attribute and Relationship .

By drawing ER Model , We have clarified the business logic , Put the drawn ER Convert the model into a specific data table , The principle of conversion is as follows :

• (1) An entity is usually converted into a data table ;
• (2) One many-to-many , It is usually also converted into a data table ;
• (3) One 1 Yes 1, perhaps 1 Relationship to many , It is often expressed by the foreign key of the table , Instead of designing a new data table ;
• (4) Attribute is converted to the field of the table .

have access to PowerDesigner Tools Draw data conceptual data model and physical data model , And it can also transform conceptual data model and physical data model , You can also export physical data models as executable sql Script .

4. Design principle of data sheet

• “ The principle of three less and one more ”

1. The fewer tables, the better
   RDBMS At the heart of this is the definition of entities and connections , That is to say E-R chart (Entity Relationship Diagram) , Fewer tables , The simpler the design of proof entities and connections , It's easy to understand and operate .
2. The fewer fields in the data table, the better
   The more fields , The greater the possibility of data redundancy . The premise of setting a small number of fields is that each field is independent , Instead, the value of a field can be calculated by other fields . Of course, a small number of fields is relative , We usually balance data redundancy with retrieval efficiency .
3. The fewer fields in the data table, the better
   The primary key is set to determine uniqueness , When the uniqueness of a field cannot be determined , We need to use the method of joint primary key ( That is to define a primary key with multiple fields ). The more fields in a federated primary key , The larger the index space , Not only will it make understanding more difficult , It also increases the running time
   Index space , Therefore, the fewer fields in the union primary key, the better .
4. The more primary and foreign keys you use, the better
   Database design is actually to define various tables , And the relationship between the various fields . The more of these relationships , Prove that the lower the redundancy between these entities , The higher the utilization . The advantage of this is that it not only ensures the independence between data tables , It can also improve the utilization rate of correlation between each other .

“ Three less and one more " The core of the principle is Simple and reusable . Simple means using fewer tables 、 Fewer fields 、 Fewer union primary key fields to complete data table design . Reusability is through primary keys 、 The use of foreign keys to enhance the reuse rate between data tables . Because a primary key can be understood as the representative of a table . The more keys are designed , Prove that the higher the utilization ratio between them .