Data Model is characterized by dataabstractIt describes the static characteristics, dynamic behaviors and constraints of the system from an abstract level, and provides an abstract framework for the information representation and operation of database systems.The data model describes three parts: data structure, data operation and data constraint[1]。
The model can reveal the essential characteristics of things more vividly and intuitively, so that people can have a more comprehensive and in-depth understanding of things, which can help people solve problems better.Using models to describe things is a widely used method in the process of understanding and transforming the world.The computer cannot directly deal with the objective things in the real world, and the database system uses computer technology to manage the objective things, so it is necessary to abstract and simulate the objective things to establish a data model suitable for the management of the database system.Data model is the simulation and abstraction of data characteristics in the real world[2]。
The data model isDatabase designThe tool used to abstract the real world in the database is a formal framework used to provide information representation and operation means.Data model is the core and foundation of database system[3]。
Components
Announce
edit
data modelThe content described includes three parts: data structureData operation, Data constraints[4]。
① Data structure: the data structure in the data model mainly describes the type, content, nature of data and the relationship between data.Data structure is the foundation of data model, and data operation and constraint are based on data structure.Different data structures have different operations and constraints[4]。
② Data operation: the data operation in the data model mainly describes the operation type and operation mode on the corresponding data structure[4]。
③ Data constraints: the data constraints in the data model mainly describe the syntax, semantic relations, constraints and dependencies between the data within the data structure, as well as the rules of dynamic data changes, to ensure the correctness, effectiveness and compatibility of data[4]。
Hierarchy Type
Announce
edit
Data models are divided into three types according to different application levels: conceptual data model, logical data model, and physical data model[5]。
Conceptual data model
Conceptual data model(Conceptual Data Model)user, forobjective worldThe model ofconceptualizationStructure, which means that the database designer gets rid of the specific technical problems of the computer system and DBMS at the initial stage of design, focuses on analyzing data and the relationship between data, and has nothing to do with the specific database management system (DBMS).Conceptual data model must be replaced by logical data model to be implemented in DBMS[5]。
The most commonly used in conceptual data models areE-R model, extended E-R modelobject-orientedModel and predicate model[5]。
Logical data model
Logical data model(Logical Data Model) is a database system oriented model, which is the data model supported by specific DBMS, such as the Network Data ModelHierarchical data model(Hierarchical Data Model), etc.This model is both user oriented and system oriented, and is mainly used for the implementation of database management system (DBMS)[5]。
Physical data model
Physical data model(Physical Data Model) is a model for computer physical representation, which describes the dataStorage mediumThe organizational structure of theDBMSRelated to, and also related tooperating systemandHardwareofEachLogical data modelThere are correspondingPhysical data model。To ensure its independence, DBMSPortabilityMost of the implementation of physical data model is automatically completed by the system, while the designer only designsIndexes、gatherAnd other special structures[5]。
classification
Announce
edit
In the process of data development, there have been three basic data models: hierarchical model, mesh model and relational model.These three models are named according to their data structures.The first two adoptformatStructure of.In this kind of structure, entities are represented by record type, and record type is abstracted as vertex of graph.The connection between record types is abstracted as the connection arc between vertices.The entire data structure corresponds to the diagram.The basic structure of the hierarchical model istree structure；The basic structure of the mesh model isUndirected graph。The relationship model is unformatted, and uses a single2D tableA relational model is a data model commonly used in databases at present[5]。
hierarchical model
Organize data into a one to many relationship structure, and use a tree structure to represent entities and their relationships[5]。
Mesh model
Using connection instructions or pointers to determine the network connection relationship between data is a multi to multi type data organization mode[5]。
relational model
The data is organized in the form of record groups or data tables to facilitate the storage and transformation of the relationship between various entities and attributes, without hierarchy or pointer. It is a very effective data organization method to establish the relationship between spatial data and attribute data[5]。
requirement
Announce
edit
The data model should meet three requirements: first, it can better simulate the real world;Second, it is easy to be understood;Third, it is easy to realize in the computer.It is still difficult for a data model to meet these three requirements well and comprehensively.Therefore, different data models should be used in the database system for different use objects and application purposes.Just as different drawings are needed at different stages of architectural design and constructionDatabase application systemYou also need to use different data models in[6]。
development
Announce
edit
1. The first generation database system hierarchy and mesh database management system
The representative product of hierarchical and mesh databases is the hierarchical model database management system developed by IBM in 1969.Hierarchical database is the pioneer of database system, while mesh database is the foundation of database concept, method and technology.
2. The second generation database system relational database management system (RDBMS)
In 1970, EF. Codd put forward the relational model of database in the paper entitled "Relational Model of Large Shared Database Data", which laid a theoretical foundation for relational database technology.By the 1980s, almost all newly developed database systems were relational.James Gray is the key person who really makes the relational database technology practical.Gray has played a key role in solving major technical problems such as how to ensure the integrity, security, concurrency of data, and the ability of database to recover from failures.The emergence of relational database systems has promoted the miniaturization and popularization of databases, making it possible to configure database systems on microcomputers.
3. Research and development of new generation database technology
At present, the current database system technology has been developed from many aspects.We can summarize the development of the new generation database system from three aspects: data model, new technology content and application field.
(1) Object oriented methods and technologies have the most profound impact on database development
In the 1980s, the emergence of object-oriented methods and technologies had a profound impact on various fields of computer, including programming languages, software engineering, information system design and computer hardware equipment, and also brought new opportunities and hopes to database technology facing new challenges.Database researchers learned from and absorbed object-oriented methods and technologies, and proposed an object-oriented database model (referred to as the object model).At present, many researches are based on the existing achievements and technologies of database. Aiming at different applications, traditional DBMS, mainly RDBMS, is expanded at different levels, such as establishing object relational (OR) model and object relational database (ORDB).
(2) Organic combination of database technology and multidisciplinary technology
The organic combination of database technology and multidisciplinary technology is an important feature of current database development.The development of other emerging technologies in the computer field has had a significant impact on database technology.The combination and mutual penetration of traditional database technology and other computer technologies make new technical contents in the database emerge in endlessly.Many concepts, technical contents, application fields and even some principles of database have undergone significant development and changes.A series of new databases have been established and implemented, such as distributed database, parallel database, deductive database, knowledge base, multimedia database, mobile database, and so on, which together constitute the database family.
(3) Research on database technology for special application field
In order to meet the requirements of diversified database applications, on the basis of traditional databases and in combination with the characteristics of various specialized application fields, research on database technologies suitable for this application field, such as engineering databases, statistical databases, scientific databases, spatial databases, geographic databases, Web databases, etc., which is another important feature of current database technology development.At the same time, the database system structure has also developed from the centralized structure of the host/terminal to the distributed structure of the network environment, and then has developed into a two-layer, three-layer or multi-layer client/server structure, as well as a browser/server in the Internet environment and a dynamic structure in the mobile environment.A variety of database structures meet the needs of different applications and adapt to different application environments.[7]
