Characteristics of Database Management System

 The following are some of the characteristics of a Database System:

1. The self-describing nature of the database system allows meta data to be stored in the database
2. Provides insulation between programs and data by providing data abstraction
3. Supports multiple views of data stored in a database at the view level
4. Allows data to be shared among multiple users in a multi-user environment

Here is a list of functions performed by a Database Management System:

1. Provides persistent storage for data processed by the application programs
2. Restricts unauthorized access from accessing the database
3. Represents complex relations among data stored in a database
4. Controls redundancy of data in a database by enforcing integrity constraints on data to be stored in a database
5. Provides efficient query processing for speeding up the retrieval of data from the database
6. Manages backup and recovery of data during system failure
7. Supports multi-user view on data through data abstraction

Three Level of Data Abstraction in a Database

The major purpose of a database system is to provide its users with an abstract view of data stored in a database.  Abstraction hides the details such as the underlying storage mechanism used for storing the data physically or logically.

A Database System provides abstraction on data stored in a database in three different levels namely:

• Physical level
• Logical level and
• View level

Physical Level of Data Abstraction:

The physical level of abstraction describes the details such as low-level data structure used for actual storage of data in memory and the mechanism involved in storing and retrieving the data physically on disk using the data structure.

Logical Level of Data Abstraction:

At logical level, data are organized in a higher-level using the conceptual structure like table, which is also called as entity or relation.  Each and every table determines what data are to be stored in a database, without exposing the mechanism used for storing the data physically in a file.  Although storage of each and every table at the physical level involves complex physical-level structure, users at logical level don't need to be aware of physical implementation of database.

View Level of Data Abstraction:

This is the highest level of data abstraction in a database.  In this level of data abstraction, the user can only view the data stored in one or more tables of a database.  View level allows the user to only have a look at the data without having the knowledge of the conceptual or physical structure of data at the lower levels.  Thus, by providing abstraction at various levels of data storage, database systems hide many of the low-level storage details from database users.

Three Schema Architecture:

The description of a database is called the database schema, which is specified during database design and is not expected to change frequently.  The three-schema architecture of a database defines the database in three different level and provides data abstraction in three different levels namely:

1. Internal or Physical level
2. Conceptual or Logical level
3. External or View level

The below figure shows the three-schema architecture of a database that describes the database in three different levels:

The internal level describes the schema of the database at physical level by describing the phyical storage structure and the mechanism used for storing and retrieving the data physically on a storage medium like hard disk.   

The conceptual level describes the logical schema of the database at the logical level by describing the structure of the database using data types, relationships, and constrains to be applied on data to be stored in a database.

The external or view level desribes the various view of the user at the application level.  At this level of data description, data are allowed to be viewed by the users at various angles.  User are allowed to view a part of the database and the rest of the database is hidden from them.  Both conceptual schema and the physical schema of the database are abstracted from the users at the view level of the database.

Data Independence:

Data independence defines the capacity of a database system to change the schema at one level without having to change the schema at the next higher level.  Data independence allows change to take place in schema at low level, without having a change in schema at next higher level of the database; only the mapping between the two level is changed.  There are two types of data independence namely, logical data independence and physical data independence.

The logical data independence is the capacity to change the conceptual schema without having to change external schemas or application programs.  A DBA may change the conceptual schema of the database for any one of the following:

1. To expand the database by adding new fields or imposing new constraints on existing fields
2. Modifying the data type or size of existing fields and changing the constraints of existing fields
3. Reducing the database by removing one or more data fields

The physical data independence is the capacity of a database to change the internal schema without having to change the conceptual schema.  The level of independence allows the physical structure of the database to be modified without requiring a change in the logical structure of the database which includes tables, relationships and constraints.