Three Level Architecture of DBMS
According the approach of ANSI/SPARC (American National Standard Institute / Standards Planning and Requirements Committee) there are three level of architecture followed in DBMS: external level, conceptual level and internal or physical level.
1. External Level of Three Level Architecture
It is the highest level of the abstraction of data. Each external level is defined by means of an External schema. Which consists basically definitions of each of the various types of external record found at that external level.
The external level is that the view that the individual user of the database has. It determines How the data LOOK LIKE, irrespective of how they are stored. It represents a part of the database. The same database may provide different classes of users.
2. Conceptual Level of Three Level Architecture
It is the next lower level of abstraction. The conceptual level is defined by means of the conceptual schema. This is also known as user’s view that describes WHAT type of data is actually stored. It simply contains the information about structure (or type) of data stored in database. A database administrator, who must decide which information, is to be kept in the database, what are the relations among the tables at conceptual level. The conceptual schema is defined using Data Definition Language (DDL).
3. Internal or Physical Level of Three Level Architecture
It is the lowest level of the data abstraction. The Physical level is defined by the means of Internal schema.
It is concerned with the physical storage of the information means what data are stored and how. It also specifies what indexes exists, how stored fields are represented. It gives complete internal structure of storage and their access paths specification of primary, secondary keys, indexes, pointers and sequencing, data compression and optimization techniques.
Mapping Between the Three Levels
Mapping corresponds one record of one level to record of another level. There are two mappings in three levels architecture.
1. Mapping between External and Conceptual level.
2. Mapping between Internal and Conceptual level.
3. Internal to physical database.
There is not necessary that an internal record at internal level is stored at a physical record, it can be from multiple records on storage device. It provides the relevant mappings.