Normalising a database

[Diellza]

Hello my friends, I am just pursuing an ordinary course in Information technology, but later on I am planning to go for a database relevant course from either SQL or Oracle, but right now I have got some sort of doubts in the simper things as well , my question is regarding the Normalization f a table in he detail, can anyone please explain it in brief.

[Vandam]

Normalisation is a procedure whereby the relations in a database are optimized to take away the prospective for redundancy. Two main problems might happen if this is not done:

• Redundant data makes a database bigger.
• Numerous instances of the same values make maintaining the data more tricky and can make anomalies.

I'm not going to expand the proposal of precise anomalies much here, but to for a short time show an update anomaly envision the problem of updating a customer address. If there are multiple instances of that customer in a database, some query designed to update the address will arwe obliged to update them all. That may look simple - as we have seen in workshops, SQL is fairly proficient of updating information that are based on field values. But if one of those column values (say an instance of student name) has been input wrongly, then the query would not succeed to update the address in that tuple.

[Dwarner]

This is perhaps most with no trouble understood as there cannot be redundant columns (that which contain the same type of data and values). For instance a cars field may have a description about a list of Ford cars named in te round brackets (Focus, Mondeo, Puma). As we know we should also formulate this field atomic, but dividing the field into any three columns into three (cars1, cars2, cars3) to grip these values isn't an alternative this time as that is going to generate e three domains that holds similar data.

[Ronchi]

Normalization targets to get rid of this redundancy by applying rules in a sequence of phases, dividing relations and generating relationships between exclusive identifiers (keys), to make certain that the database table structure is well-organized, but data can still be precisely manipulated. It can also be used in combination with other database modeling techniques, such as Entity-Relationship diagrams, but for uncomplicated databases normalization can sometimes be adequate.

[Steyn]

But before you begin with the Normalization techniques you should clearly understand the Relational data model correctly , this thing are really important , if you want to pursue your career in this fields and then turn out to be a Data base administrator, the relational model imlies that all the relation are expressed in the tabular format and each table is connected or linked with a primary key foreign relationship. If you really want to learn this you can make the use f the tool MS SQL 2005 version.