1.1. Stages of database design

Printer-friendly version

The design of a database is not a simple process. Typically, the complexity of the information and the number of requirements of information systems make it complicated. Therefore, when designing databases is interesting to apply the old strategy of divide and conquer.

Therefore, it should break down the design process in several stages at each intermediate result is obtained that serves as a starting point for the next stage, and the last stage we have the desired result. Thus no need to solve at once all the problems inherent in the design, but in each stage, facing a single type of subproblem. This divides the problem and at the same time, it simplifies the process.

Decompose the design database in three stages:

The result of conceptual design

If we take the idea of three worlds, we can say that the conceptual design stage yields a result that is in the world of representations.

1) conceptual design stage: this stage we have a structure for future information technology independent BD to be employed.Not yet taken into account what type of database used-relational, object oriented, hierarchical, etc .- thus not taken into account how DBMS specific language or how it will implement the database.Thus, the conceptual design stage allows us to concentrate solely on the problem of structuring information, without having to worry about at the same time to solve technical issues.

How to develop a conceptual design expressed in the ER model described in paragraph 2 of this unit.

The result of the conceptual design stage is expressed by a model of high-level data. One of the most widely used is the model entidadinterrelación (entity-relationship), which abbreviated with the initials ER.

The result of logical design

The result of logical design is now at the world of representations.

2) Logical Design Stage: This stage is part of the result of conceptual design, which is changing to fit the technology to be employed. More specifically, it is necessary to fit the model of the DBMS with which you want to deploy the database.For example, if it is a relational DBMS, this stage will get a set of relationships with attributes, primary keys and foreign keys.

This stage of the fact that it has solved the problem of structuring information in a conceptual level, and can focus on technological issues related to the model database.

The logical design of a relational database described in paragraph 3 of this unit.

Later he describes how the logical design of a relational database, taking as its starting point a conceptual design model expressed with the ER, that is, see how you can transform an ER model into a relational model.

The result of the physical design

The result of the physical design stage is in the world of representations, as the result of the logical design stage.
The difference from the previous step is now taken into account aspects of a more physical world of representations.

3) physical design stage: This stage transforms the structure obtained in the logical design stage, with the aim of achieving greater efficiency, in addition, complete with physical implementation issues will depend on the DBMS.

For example, if it is a relational database, transformation of the structure may consist of the following: Accounts and any relationship that is the combination of several relationships that have been obtained in the logical design stage, from a relationship number, add a calculable attribute to a relationship, etc.The physical implementation issues that must be completed normally consist of the choice of physical implementation of relationships, the selection of the size of buffers (buffers) or pages, etc..

How to develop a conceptual design expressed in the ER model described in paragraph 2 of this unit.

The result of logical design

The result of logical design is now at the world of representations.

The logical design of a database
relational data is explained in
paragraph 3 of this teaching unit.

In the physical design stage in order to achieve good performance of the database, "should take into account the characteristics of the processes that query and update the database, such as access roads using and frequency of execution.It is also necessary to consider the volumes that are expected to be of different data to be stored. </ P>


Datawarehouse loadings are periodic, and summary tables are useful to improve efficiency and the answer time of our reports. A physical optimisation resource that can bring big improvements in our system is the utilization of materialized views. The view materialized is not more that a view, defined with one sentences SQL, of which in addition to storing his definition, store the data that...
We've completed the publication of the Tutorial found RapidMiner 5.0 , one of the best Open Source tools for data mining, which also has Enterprise versions. This tutorial shows the basics of RapidMiner and simple process configurations can be performed.It shows how to perform each datamining process with an example that details the steps. The second part of the tutorial is a...
The news of Cloud Computing are continuous, as a model that every day gets better acceptance, because of its many advantages. Among those is, as already discussed at other times, the cost savings in both software and hardware, the advantage of not having to make investments in equipment or safety features and availability of data. For those who are interested in a deeper approach to cloud...