DB Concepts at General Connection dot Com Logo
N A V I G A T I O N
Home
Database Terminology
All about DBMS
Whats New?
 
Relational DBMS and Key Components


RELATIONAL Database Management Systems

  1. Eliminates the need for explicit parent-child relationships
  2. Represents relationships by two-dimensional tables i.e in an RDBMS, data is perceived by the user as tables (and nothing but tables)
  3. Has got no physical connections between relations, i.e, no pointers are maintained between the tables


What is a Table ?

  1. A table closely resembles a conventional sequential file
  2. A row, usually referred to as tupie, is similar to a record in the file system
  3. A column, usually referred to as attribute, is similar to a field in the file system
  4. Table is an entity with a unique name, made up of rows and columns

  5. Each column has a unique name

  6. A particular row is identified by a column or a combination of columns

  7. All data values are atomic (i.e, there exists only one value, and never a set of multiple values, at a row-column intersection) 
  8.  Each table has a unique identifier

 

Customer Table

 
account-no address phone customer name
100 austin, tx 6676767676 Jerry
130 chicago, IL 4564564545 Sharon
150 NewYork, NY 1232342323 Richard
       


Ledger Table

 
account-no balance
100 10000
130 20000
150 8000

 

Notice that the ACCOUNT-NO appears in both the CUSTOMER and LEDGER tables. This column is used to establish relation between the two tables

What is a Domain?

  1. It is a pool (or range) of values from which the actual values appearing in a given column are drawn

  2. Each column can have a domain specified for it. As an example ... in the LEDGER table, ACCOUNT-NO is specified to have a domain 100 to 150
  3. Relationships between entities are established by common columns containing identical values from a domain or range of values

 

What is a Candidate Key
Also referred to as Unique Identifier or Key .
A table can have multiple candidate keys
Column or a minimum combination of columns, used to identify uniquely the tuples of that table

What is a Primary Key
Column (or a combination of columns) used to uniquely identify a tuple
Primary Key a single candidate key used to identify a unique tuple
An Example ... ACCOUNT-NO can be referred to as the primary key of the LEDGER table

CUSTOMER-NAME can be referred to as the primary key of the CUSTOMER table

Composite Key or Concatenated Key

It is a combination of candidate keys used to identify a unique tuple

example : Combination of CUSTOMER-NAME and ADDRESS, if CUSTOMER-NAME doesn't contain unique values, can be referred to as the composite key of the CUSTOMER table

What is a Super Key

Set of one or more attributes which includes one of the candidate keys, taken collectively.to identify uniquely a tuple of the table
Example ... Combination of CUSTOMER-NAME and ADDRESS, can be referred to as the super key of the CUSTOMER table

In a candidate key, there exists only one column or a minimum combination of columns , while, in a super key there is no such restriction

Sometimes the super key may be the candidate key itself

A candidate key differs from a super key

What is a Alternate Key

Column or combination of columns KEY which has unique values but not selected as a primary key and is not a part of a primary key

That is to say... A candidate key that can be a primary key but is not so

What is a Foreign Key

Eg ACCOUNT-NO in the CUSTOMER table can be referred to as foreign key, it being a prinialy key in the LEDGER table
Column (or combination of columns) of a table, used to establish relationship with other table(s) and present itself in all the related tables

The values of the foreign key are required to match the values of the primary key(s) in the related tables in order to enforce referential integrity

What is Referential Integrity

By definition, a foreign key in one table matches a primary key in another table

A primary key value must be unique, and cannot be null A foreign key value must match a primary key or null

INSERT/UPDATE of foreign key must match primary key of that table (where it exists) or be null

UPDATE of primary key is not allowed if any matching foreign key (in that table) exists

What is a Secondary Key

Column (or combination of columns) which does not identify a unique tuple but, which identifies all those tuples which have a certain common property

example... PHONE and/or ADDRESS can be referred to as the secondary key of the CUSTOMER table

Hierarchic or Network structures deal with one record at a time i.e, data is read/written in units of a record as in the familiar file management systems

In an RDBMS, interaction is in units of a set of records and these interactions are based on the mathematical notions of relations (relational calculus)







© 2010 www.generalconnection.com. All rights reserved. | www.generalconnection.com