Soft test Summary (1)--Database design

Database design, is a big problem in the test, of course, is also our development process is essential, and now a summary of the database design.

first, the database design phase Division

First of all, to take a picture of the database design process has a visual understanding

(1) Demand Analysis
Objective: To understand the requirements and processing requirements of users through research and investigation .

Results: To form a requirement specification. Basis and basis for future design. Include: Data contained in a database, characteristics of data, frequency of use and data

volume of estimates, such as data names, types, requirements for data, and size of usage, are typically used in data dictionary management

(2) Conceptual structure design
Objective: To abstract the data and requirements of demand analysis, to construct the data that reflects the user's environment, and the conceptual model of the interrelated data.

(3) Logical Structure design
Objective: To convert the E-r diagram into a relational pattern according to the relevant rules. Determining the characteristics of primary key, foreign KEY and constraint according to the theory of normalization

Result: The data definition language provided by the DBMS, the data pattern written

(4) Physical design
Objective: The logical model is transformed into a physical model, including the design process of file structure, index, etc.

Second, e-r model

(1) Includes: entity, attribute, contact, type of contact (number)

(2) Design steps: First design the local e-r diagram, then merge (view integration). Eliminate inconsistencies between local er graphs when merging

Conflict types for local ER graphs when merging:

1) attribute conflicts: Conflicts of attribute domains and violation of attribute values

2) Naming conflicts: With the same name, and with the consent of the different names

3) Structural conflicts: different abstractions for the same object in different applications, the same entity is not exactly the same as the number of attributes


(3) Local e-r diagram design steps: Determine the scope of the local view (for example, related to procurement), identify the local entity and the

Contact Anti- feed entities and the properties of relationships between entities

E-r diagram-to-relationship mode conversion:

1) converting an entity to a relational schema

2) A one-to-one relationship is transformed into an independent relational pattern-attribute: The code of the contact itself and the attributes associated with it with any one

end corresponding to the relationship mode merge--attribute: Add 21 relationships to the code and the properties of another relationship

3) A one-to-many relationship. You can form a single relational pattern-the code of individual entities and the properties of the contact itself. merge with N-end--

Add the code of the other end and the properties of the contact itself

4) Many-to-many links. can only be converted to a separate relational pattern. The code of each entity connected to the contact and the properties of the contact itself

5) Contact to connect multiple entities. Similar to the first four points

Soft test Summary (1)--Database design