Home
Designing databases
The first step in creating a relational database is to construct a data model, which is a precise, complete definition of the data you want to store. After you prepare your data model, you must implement it as a database and tables. To implement your data model, you first select a data type for each column and then you create a database and tables and populate the tables with data. You can also implement fragmentation strategies and control access to your data.
Designing and Implementing a Database
The Informix® Database Design and Implementation Guide provides information to help you design, implement, and manage your Informix databases. It includes data models that illustrate different approaches to database design and shows you how to use structured query language (SQL) to implement and manage your databases.
Basics of database design and implementation
Select data types
Define the domains

Designing databases
The first step in creating a relational database is to construct a data model, which is a precise, complete definition of the data you want to store. After you prepare your data model, you must implement it as a database and tables. To implement your data model, you first select a data type for each column and then you create a database and tables and populate the tables with data. You can also implement fragmentation strategies and control access to your data.
- Designing and Implementing a Database
  The Informix® Database Design and Implementation Guide provides information to help you design, implement, and manage your Informix databases. It includes data models that illustrate different approaches to database design and shows you how to use structured query language (SQL) to implement and manage your databases.

Define the domains

To complete the data model that Build a relational data model describes, you must define a domain for each column. The domain of a column describes the constraints and identifies the set of valid values that attributes (columns) can assume.

The purpose of a domain is to guard the semantic integrity of the data in the model; that is, to ensure that it reflects reality in a sensible way. The integrity of the data model is at risk if you can substitute a name for a telephone number or if you can enter a fraction where only integers are valid values.

To define a domain, specify the constraints that a data value must satisfy before it can be part of the domain. To specify a column domain, use the following constraints:

Data types
Default values
Check constraints
Referential constraints