What is Data Independence

Data Independence

The separation of data and application program is called data independence. It is an important feature of DBMS. It is the most important advantage of the three-level architecture. Another major advantage is that any change in the lower level of three-level architecture does not affect the structure or functionality of upper levels.

Data independence enables the user to change the structure of a database without changing application programs or the way users access the data.
There are two types of data independence:

1. Physical data independence
2. Logical data independence

 

Physical Data Independence
Physical data independence is a type of independence that enables the user to change the internal level without changing the conceptual level. In a DBMS, the physical structure of the database may change without changing application programs or altering the user's view of data. It is possible because DBMS uses abstraction. Data is translated from the way it is physically stored on a disk to the representation and access techniques used by the logical view.
               If the physical structure changes, DBMS is aware of these changes but still provides the same logical view. The logical view remains constant and the application programs and user interactions based on the logical view of data are not altered. The changes that may be performed at the physical level without changing the logical level are as follows:

1. Changing file organizations or storage structures

2. Using different storage devices

3. Modifying indexes

4. Modifying hashing algorithms

5. Changing the access method

 

Logical Data Independence
Logical data independence is a type of independence that enables the user to change the conceptual level without changing the external level.

Some additional information may be added to the database by changing its logical structure. This change should not affect user interaction or application programs. This is known as logical data independence. The changes that may be performed at the logical level without changing external level are as follows:

1. Addition or removal of entities or relationships

2. Adding a file to the database

3. Adding a new field in the file

4. Changing the type of a field etc.

          In some situations, a change that may look similar to the above-mentioned changes can create a problem. Suppose an attribute is deleted from the database structure. It is serious because any application that is using this attribute may not run anymore. It is important to analyze the effects of a change before making that change to the database.
 

• Database Environment