Normalization can be described as a breaking up of the information into
multiple parts or logical units in the form of tables. The concept of
normalization is used for organizing the data in database in a managed manner
where the redundancy of data does not happen & data remains accurate.
Normalization is basically a process of reducing redundancy
i.e. Data repetition in database for similar type of information where-in
De-normalizations only limited number of tables are created. There are lot of
many more merits & demerits of Normalization & De-normalization so
let’s look them.
A non-normalized database may contain some tables which may
contain the most of data into a few tables. There are some points which
indicate the general problems of non-normalized database.
1.
Data is stored in few tables only instead of multiple
tables but this is not good as data is easily identical which is bad practice
from the security perspective
2.
Query
performance will be affected as the table will grow drastically
3.
Activities like Update will be affected causing
the poor efficiency of database and Data integrity.
In a process of applying normalization on database few points
needs to be keep in mind among which some are as follow,
·
Data should not be redundant (Duplicate) or very
minimum for some cases for ex. In a candidate management system where
candidates information is stored. Here in this case the address & other
optional information is not required to stored in a same table as it is not
always required while searching basic information about candidate.
In the process of normalization naming convention is always
an important point because in vice-versa to de-normalization the data is split
into multiple tables & during the needs the tables needs to be joined with
one another & in this case the identical & proper naming convention will
play a key role.
Normalization benefits the database in many ways some of
them are as follow,
1.
Proper database organization
2.
Avoiding redundant data
3.
Database consistency and integrity in a which
assures the accurate & co-related data within same database.
4.
Proper & Secure database system as an
unknown & unfamiliar user cannot find & understand data easily.
5.
Referential integrity means the values of one
table is dependent upon the one or more than one number of columns
6.
Having a parent child relations between tables
so that any referential data is not lost
As like the other thing normalization also has some
drawbacks which is that due to heavy normalization during the processing of
data during the insert, update or select all the relational objects are
required to be bring in use which cause the high CPU use & IO operation increase
so although the huge benefits the normalization should be carried out wisely.
De-Normalization:
Almost described most of the differences between normalization
& de-normalization & the various pros & cons of both concepts
Description of de-normalization in simple terms “Instead of
storing data into multiple space a few number of tables where most of the data
can be stored for stress free environment without caring about too much of data
hiding methodology”
No comments:
Post a Comment