- What we have learned so far:
- The Physical Data Independence
(meta data) technique:
- This programming technique allows us to write one single program to access all possible data files (each one with different data format)
- The Logical Data Independence
technique:
- This technique uses a set of programs that allow us to combine (= merge) different data files and present the data in a format/form desired/required by different users
- The Physical Data Independence
(meta data) technique:
- Bold Claim:
- By
using these 2 techniques
- the Physical Data Independence technique, and
- the Logical Data Independence technique
we can develop one (huge) computer program (called a Database Management System (DBMS) that can satisfy the information needs of every groups of users
- In other words:
- We can use
one (huge) computer program
to:
- store the data files efficiently
- and yet, be able to present the data to any user in the precise format that the user want to see
- We can use
one (huge) computer program
to:
How to accomplish this:
- We collect
data (information) need from
every (group of) user
as views on the
data:
The views are the external schemas that we must support
- We design
a (one) conceptual schema
and a set of (SQL) programs to
map the
conceptual schema to
the external schemas:
(This is Logical Data Independence !!)
- We store the
data files
(on disk) and
define
the meta data for the
data files
(so the database program can
access the
data files):
(This is Physical Data Independence !!)
- By
using these 2 techniques
- The one piece of
(huge) computer program that
I am alluding above --- is known as:
- a database management system (DBMS)
- The following figure shows the internals of
a DBMS system:
The DBMS software is the part inside the dark green region
- Operation:
- When a user (say user 1)
wants to view his data, then:
- User 1 view program is run
(This is a very small program written in SQL (discussed later) that is stored inside the DBMS)
- User 1 view program is run
- When the DBMS system
runs the
User 1 view program, it will
need to access various
data files
- In order to access the
different files, the DBMS first reads the
Meta Data
- Using the Meta Data information, the DMBS then access the data files that are required to construct the data desired by user 1
- In order to access the
different files, the DBMS first reads the
Meta Data
- The result is then formatted (by the User 1 view program) into the desired format as requested by the user
- When a user (say user 1)
wants to view his data, then: