What is the simple definition of the Normalization in the Relational Database Management System (RDBMS) like SQL-Server and Oracle? What is the main difference in the terms called “Normalization” and “Denormalization”?
Here is always a confusion to understand these two very important terms called “Normalization” and “Denormalization” in the database like SQL-Server, MySQL and ORACLE.
There are some principles to follow in database to get the better performance and clarity of understand the schema. These principles are known as “Principles of Normalization”. At present we are having 1NF to 6 NF (NF means Normalization Form). As highest we go we will get the better performance of the server to update the data, insert the data and delete the data at less cost.
The definition of the Normalization is to divide the very large table into two there small tables to reduce the Redundant Data (Duplicate Data) and the dependency. If we can do that we can have the better performance and better results from our database. For example if we are going to represent the data in the one large table, then definitely we will get confused with the records in the given table. The simple reason is that we are going to see the duplicate data and lot of dependencies among the fields in the given table.
Then we need to split the data into the small tables to reduce the Redundancy or Duplicity and dependency among the fields. This process is called the “Normalization”. This is really a very good term in the database to create the database in the useful manner, which is better than the creating the mess in the big database.
If we are good enough in creating the relation in the Entities and Attributes or in simple language we can say that Tables and Columns. Then we are good to go with ease to handle the big data base.
De-Normalization is opposite to the Normalization because it support the “Star Schema” and Normalization support the “Snow Flake Schema”. “Star Schema” is useful in the reading the data because of the less number of the tables. In “Snowflake Schema”, numbers of tables are more as compare to the “Star Schema”.
If we are working as Database Architect, then it is must to know these processes of Normalization and De-Normalization based on the “Snowflake” and “Star” Schema. Then we can create an understandable database. Otherwise we will not be able to solve our purposes.
There is one more fundamental fact for designing the database for multiple problems is that we can use the 3rd Normal form (3NF) to maintain the database structure without any change to our database to solve the problems.
I am trying to explain it in more detail in next coming blogs. It is not easy to understand this concept but once you fully understand this concept then you will be good enough to design the database for the big companies.
I request expert in this field to have some comments to explain it in detail for the sake of more knowledge and experience.
Thanks,
Sandeep Rajmail
Leave a Reply