Basic steps of Database design __ Database

Basic steps of database design

According to the standard design method, considering the whole process of database and its application system development, the database design is divided into the following 6 stages

1. Demand Analysis

2. Conceptual structure Design

3. Logical Structure Design

4. Physical Structure Design

5. Database implementation

6. Operation and maintenance of database

In the process of database design, the requirement analysis and conceptual design can be carried out independently of any database management system, and logical design and physical design are closely related to the dams of selection.

1. Requirements analysis phase (often from top down)

Database design must first accurately understand and analyze user requirements (including data and processing). Requirements analysis is the basis of the entire design process, but also the most difficult, most time-consuming step. Whether the requirement analysis is done adequately and accurately determines the speed and quality of building the database building on it. The need analysis does not do well, will cause the entire database design rework redo.

The task of requirements analysis, is through a detailed investigation of the real world to deal with the object, fully understand the work of the original system, clarify the needs of users, and then determine the new system functions, the new system must fully consider the possible future expansion and change, not only according to the current application requirements to design.

The focus of the survey is on data and processing. Meet information requirements, process requirements, security, and integrity requirements.

The analysis method is commonly used to analyze the structure of SA (structured analysis), the SA method is based on the top layer system structure, and the system is analyzed by top-down and layered decomposition.

The data flow diagram expresses the relationship between the process and the processing, in which the processing logic of the process is often described by the decision table or decision tree. In the process of decomposition of the function of the colleagues, the system of data is also gradually decomposed into a number of levels of flow diagram. The data in the system is described by means of a data dictionary (DICTIONARY,DD). Data dictionary is a collection of various data descriptions in a system, which usually includes 5 stages of data item, structure, data stream, data storage, and processing process.

2. Conceptual structure design phase (commonly used from the bottom up)

Conceptual structure design is the key of the whole database design, and it has formed a concept model which is independent of the specific DBMS by synthesizing, summarizing and abstracting the user's requirements.

The design concept structure usually has four kinds of methods: Top-down. That is to define the framework of the global conceptual structure first, and then refine it gradually. From the bottom up. The concept structure of each local application is defined first, then they are integrated into the global conceptual structure. Gradually expand. First, define the most important core conceptual structure, then expand outwards, and then incrementally generate other conceptual structures, until the overall conceptual structure. Mixed policy. That is, Top-down and bottom-up combination.

3. Logical Structure design phase (e-R diagram)

The logical structure design is to transform the conceptual structure into a data model supported by a DBMS and will be optimized.

In this stage, the E-R diagram appears to be extremely important. You have to learn the attributes defined by each entity to draw the overall E-R diagram.

There are three main kinds of conflicts between E-R graphs: attribute conflict, naming conflict, and structure conflict.

In the transformation of E-R diagram to relational model, the problem is how to transform the relationship between entity and entity into relational pattern, and how to determine the attributes and codes of these relational patterns.

4. Physical design Phase

Physical design is to select a physical structure (including storage structure and access method) that best suits the application environment for the logical data structure model.

The first step is to analyze the running transaction in detail, get the parameters needed to select the physical database design, and, second, fully understand the internal characteristics of the RDBMS used, especially the access method and storage structure provided by the system.

There are three kinds of common access methods: 1. Indexing method, currently mainly B + Tree indexing method. 2. Cluster method (clustering) method. 3. Is the hash method.

5. Database implementation phase

Database implementation phase, the design personnel operating DBMS provided by the database language (such as SQL) and its host language, according to the logical design and physical design results of the establishment of the database, programming and debugging applications, organizing data warehousing, and trial run.

6. Database operation and Maintenance phase

The database application system can be put into operation after trial operation, and it must be evaluated, adjusted and modified continuously in the process of database system operation.