Is the concept of Denormalization an exact opposite of normalization? If not what is it? Does denormalization means redundancy of data is involved in the tables considered. Anyone describe on the topic.
Printable View
Is the concept of Denormalization an exact opposite of normalization? If not what is it? Does denormalization means redundancy of data is involved in the tables considered. Anyone describe on the topic.
Denormalization is the process of achieving optimization in database by including redundant data. But denormalization can occur only after some levels of normalization have been applied to database. In other words relying completely on denormalization is not efficient to database design.
Also I think denormalization mostly apply to logical design rather than physical design. Am I right in my statement? If not correct me!!!!
Denormalization is a data model designing technique. It is used for Data Warehouse designing. your transactional systems ([OLTP] - Billing or online booking systems for example) are in generally 3NF normalized database.
When you are designing a DWH, you turn your OLTP system which is in normalized format, to denormalized state. lets take an example,
your online booking system (OLTP) has customer information as well as say travel destinations information.
so you will have different tables for
--Country, state, City, County, Zipcodes etc.
which will be connected to each other in uniformed hirarchey.
In DWH, You create a dimension for Locations/Regions under which you will have all the data of locations together.
it [COLOR=red]means in physical table "Locations" in each row you will have Country, state, City, County, Zipcode.[/COLOR]
[COLOR=black]though values are not atomic, it serves purpose of DWH by reducing tables & hence joins & so improving performance.[/COLOR]
Denormalized database exists in phisical tables.