Introduction to database management systems

The ANSI/SPARC (American National Standards Institute/Standards Planning and Requirements Committee) suggested three level architecture for DBMS.A database is considered as containing data about an enterprise. The three levels of architecture are three different views of data:

• External - user view of  individual
• Conceptual - user view of community
• Internal - physical or storage view

The three-level database architecture allows a clear separation of information meaning (conceptual  view)  from  external  data  representation  and  from  physical  data structure  layout.  A  database  system  that  is  able  to  separate  the  three  different  views  of data  is  likely  to  be  flexible  and  adaptable.

The external level is a view which individual user of database has.  This  view  is often  a  restricted  view  of  database  and  the  same  database  may  provide  a  number  of different  views  for  different  classes  of  users.  In general,  end  users  and  even  application  programmers  are  only  interested  in  a  subset  of  database. For example, a department head will be only interested in departmental finances and student enrolments but not library information.

The conceptual view is information model of enterprise and contains view of whole enterprise without any concern for physical implementation. In a database, it may be desirable to change internal view to improve performance while there has been no change in conceptual view of the database. The conceptual view is overall community view of database  and  it  includes  all  information  that  is  going  to  be  represented  in  database. 

The  internal  view  is  view  about  actual  physical  storage  of  data.  It tells us what data is stored in database and how.  At least the following aspects are considered at this level:  Storage allocation e.g. B-trees, hashing etc. 

The  separation  of  conceptual  view  from  internal  view  enables  us  to provide  a  logical  description  of  database  without  need  to  specify  physical structures which is often called as physical data independence. Separating external views from conceptual view enables us to change conceptual view without affecting external views which is often called as logical data independence.