Overview of database system -- overview of data model

Non software major , I have always felt it difficult to convert the demand to the data of the provincial database , The data model part in the overview of database system speaks well about this part , Sort out and record .

When converting requirements into tables in the database and files on the operating system , It roughly goes through the following steps , demand ->  conceptual model ->  Data model supported by database management system ->  The physical model （ Physical file ）

conceptual model

It is used for modeling the information world , The language of communication between database designers and users , Need strong semantic expression ability , Simple 、 Clear 、 Easy for users to understand , The expression method is ER chart ,EA The tool DataBaseEngine->EntityRelationShip,

 ER The graph has entities 、 Attribute and connection are three parts

Entity ： Transactions that exist and can be distinguished from each other , It can be human. 、 things 、 Things or abstract concepts , Because it can be distinguished , There must be a unique key , Namely code

attribute ： That is, the attribute of the entity

contact ： The connection within the entity （ Not yet learned ）、 Connection between entities 1:1  1:n  m:n

  logical model

The logical model is through structure 、 operation 、 Integrity constraint expression Of ,

structure ： Describe the constituent objects of the database （ Include type 、 Content 、 nature ）, And the connection between objects （ Set up Type, That is, the ownership relationship ）, Is a static description

operation ： Add, delete, modify and check the object , Allow those , Consider querying and updating （ Additions and deletions ） that will do , Dynamic description

Integrity constraints ： A set of integrity rules , Integrity rules ： Constraints and dependency rules of data and its relationships in a given data model , It is used to define the state of the database in accordance with the data model and the change of the state , To make sure the data is correct 、 Effective and compatible , It is divided into basic general constraints and constraints that need to be used in specific applications .

The types of logical models are hierarchical model 、 Mesh model 、 relational model 、 Object oriented model 、 Object relational data model 、 Semi structured data model 、 Unstructured data model , The hierarchical and mesh models are formatted models , Use hierarchical connection description , Outdated .

hierarchical model

The hierarchical model uses A tree structure To represent all kinds of entities and the relationships between them ,XML Language , The set of basic hierarchical relations satisfying the following two conditions is a hierarchical model

Operational and integrity constraints ： Constraints can be handled with foreign keys

• Without the corresponding parent node value, the child node value cannot be inserted , Only one root node , If there is no parent node , You can't insert
• If you delete the parent node value , The corresponding child node values are also deleted
• When updating operation , All corresponding records should be updated , To ensure data consistency

shortcoming ：

• Cannot achieve many to many problems
• There are many insert and delete operations , Query child nodes can only be downward through parents layer by layer

Mesh model

Solve the single root node of hierarchy , That is, more than one node is allowed to have no parents （ Multiple root nodes ）, Solved the problem that the hierarchy cannot specify multiple parents , That is, find from multiple paths , Solve many to many problems .

  a key ： Many to many problems can be solved by decomposing into one to many connections

EA The tool DataModel chart , The following decomposes the many to many relationship between students and courses into the following one to many relationship .

  Operations and constraints ：

（1） Must have the concept of code , Because there is no root

（2） Ensure that the parent node and child node are in a one to many relationship , It is convenient to layer the network

（3） There are constraints on the relationship between parent nodes and child nodes , That is, the problem of nationality , The existence of parents is required to insert child nodes , Delete parent node , Child nodes must also be deleted , For example, course selection must rely on the existence of both students and courses to insert , When deleting a student , Students' relevant elective courses have to be deleted .

shortcoming

（1） Complicated structure , Difficult to understand

（2） Use data according to 1 Find multiple paths , You have to follow the path

relational model

A relationship is a table , Contains tuples （ a line ）, attribute （ Column ）、 Main code （ A tuple can be uniquely determined ）、 Domain （ Property value range ）.

Standard conditions ：

（1） Each component of a relationship must be an indivisible data item , It is not allowed to have tables in the table

Salary in the subscript , Including basic salary , Job pay , Performance allowances , It's a little watch , Deduction is the same small table .

  advantage ：

（1） The operation objects are all relations

（2） Hidden path , Easy to use

shortcoming ：

Access path is transparent to users , Query efficiency is often not as efficient as the formatted data model

To improve performance , The user's query request must be optimized , It increases the difficulty of developing database management system