Analysis of demand
Database design, first of all have a clear demand, throughout the design process database should always pay attention from design to final inspection data integrity constraints, including entity integrity, referential integrity integrity, domain integrity, user-defined (business rule-bound )
Data Sources
Before designing the database, start the requirements document to extract data from the database design should contain, in accordance with the understanding of the needs of data in the database There are three main sources.
First: system initialization data to import
Second: storing user data
Third: The new data handled by the software out of these two data
Data whereabouts
Output data in the database has two main destination:
First: As system data
Second: communication to the user
After learning of knowledge, started designing CDM
According to the source, I designed the corresponding entity
First: system initialization data to import
Software interface entity
Template entity
Role entity
User rights entity
Migration extent entity
Second: storing user data
User entity
Pictures entity
Migration pictures entity
Third: The new data handled by the software out of these two data
1. The definition of the relationship between the user and the collection of pictures
2. The user defines the physical history between templates, image, picture migration
Physical design field
The main design reflects the picture in terms of storage: Because the picture is relatively large, can not be stored directly in the database, so the pictures, templates, migration picture, unified storage storage path corresponding
In order to unify other data format, add a number to each non-entity relationship for unique identification. No consistent length, to facilitate the transmission and processing of data.
- Calibration
After completion of the design corresponding entities, conducted verification of CDM, mainly for the following two aspects of thinking
The data is complete
By checking, I can not judge whether the data is complete, because there are many places of uncertain demand, which led to no standard check.
Whether the data redundancy
By check, in my understanding, the design of the CDM is not the case of data redundancy exists
to sum up
Summary: Huang Kohei
First, through this database design is mainly obtained the following lessons
1. When you use a software tool, go to the learning tutorials, which will save a lot of time, also avoid weird error occurred.
2. When demand is not determined, not to design a database, which can lead to failure of database design, from the design requirements for the project did not produce results.
For database design need to draw the CDM, PDM and data dictionary
Second, the completion of preparation
Mainly through online video tutorials, comprehensive study of the CDM's design, which will not lead to lack of knowledge can not grasp the modeling software, complete modeling in the design process. From the video tutorial, I learned that the main role powerdesigner CDM portion of the tool, each tab corresponding functions, and to avoid a lot of unnecessary mistakes. E.g:
Unlike 1.CDM PDM, will not have the same field in different entities, the primary foreign key relationship expressed by the PDM
2. In my understanding, many relationships equivalent to the actual database table, if you want to add additional fields in the table, the entity should not be used to replace the relational tables in the CDM in place of relationships through association box
3. Do not occur with other entities in the same field in the association, and should be represented by such data rely on many relationship, when forwarding to the PDM, many relationship automatically a corresponding master key of the entity as a multi- foreign key to the entity.
4. CDM in the design, to be noted intentify fill, indentify the primary identifier of the CDM, a plurality of identifiers may be added, others are secondary identifiers.
In fact there are many considerations, not by learning system is not learned.
Summary: Zhang Ding
Recent talk about writing some database experience, although that is prepared for the first database, but in fact we are in this big task interspersed with a lot of small experiments, according to the experimental database of many, in fact, so our experience is in constant amount increase, from some of the basics at the beginning of the school curriculum, from time to guide the teacher behind the actual preparation of the database where some experienced guidance available, even when the database is initially written in the table are not rigorous, data type and associated infrastructure entities not OK, then we are in the preparation of the number of tables will be too little to limit the number of tables, some tables are very unreasonable, but also because it is a basic software, so it needs some relevant scalability, such as allowing users to increase the input picture number, etc., which we are beginning to limit the amount of time the user input, but actually expand disadvantage. then is the lack of consideration of the actual content in the related commentary, for example, when the history data storage we started all the pictures stored with the data, just set the relevant fields calls But historical data is actually less calls, should set up a table to store additional, new data will not be affected during this call, and in the design of relevant content insufficient use of the tools of awareness is a problem, there are also some features later, the Internet search to know, for example, rearrange the position of the table cloth, different parts of a database can be divided and writing files, which are written in the database of the harvest.
Database design is the cornerstone of a project, database design quality directly affect the progress of the post-code development, here to talk about some of the ideas and experience in the design process.
Summary: Dong Junhong
1. First, we need to understand the basic purpose of the project was analyzed according to the demand, we need about how much summed up tables, designed on the table, in fact, should not only consider only those entities on the surface of the table, should also be considered, according to some types of tables that may be invisible to the user, there is no specific form of expression, but can greatly reduce the number of unnecessary work, and to help us design the project in some of the features of 2. specific design on the database, I think needs analysis should be carried out depending on the circumstances, may demand the same, the same project, under different design concepts, can not require exactly the same requirements 3. in the design of the database, we should clarify the relationship between tables according to these relationships, and the first second and third paradigm, to standardize the database at the same time, in order to facilitate use of our database, we can consider adding some views, stored procedures, triggers, etc. as well as tools to facilitate our use, but also need to be considered when the time calls depletion
Summary: Nanjing Road
1, in the team, I appreciate the importance of communication, may accomplish something personal or a function of time need only refuel alone on the line, but we need coordinated exchange in the team.
2, the database design and database entities need to determine the demand from the field a little bit, then slowly determine links between entities and entity; it is a process of building a house of only working slowly and deliberately effort to really the complete; if we had the same experience and mature judgment that we can very quickly get the job done, so for us now take it one step at a pit;
3, the project's needs for a database is weary of goblins, let you swing stop; changes in demand due to multiple databases so that multiple rework, too weary of;
4, the database design is reasonable and practical, these really difficult to decide. Because the database is no standard answer, only better, more practical and easier to maintain. So a relatively better database needs to build wisdom of many people;
Summary: Jiang Hai with
Database design is the cornerstone of a project, database design quality directly affect the progress of the post-code development, here to talk about some of the ideas and experience in the design process.
Identify needs: be sure to carefully analyze the clear requirements and then hands-on, needs analysis inadequate or incorrect design of the database began late change will require repeated many times, time-consuming pre-designed database.
1. Determining Entity: The first is a meeting to discuss entity, since the beginning requires us to have at least 15 tables, so the initial hash table of the process is very difficult, the final confirmation with the teacher can be reduced to 10 tables.
2. Make sure the table: the general design of the table to note here is that the "relationship" in a relational database, should pay attention to the relationship between the analyzed before the distinction between the real one to one, one to many or many to many.
3. Determining fields: analysis to determine a good field in the table, including whether the field can be empty, data type, primary foreign keys, indexes, etc. is set, this step may be combined with the previous step and three paradigms of database design consideration.