Now, the question arises after identifying the facts & dimensions, how do we load them? The loading of a DW
is done by a process called Extract, Transformation & Load (more popularly known as ETL). Each & every letter in this
abbreviation has tremendous importance. Extract is a process wherein the data from a transactional system is extracted. This
data is then cleansed & transformed. Most of the applications populate the databases with information that have no meaning
from a business perspective. Such data have to cleansed/transformed to a more usable form. This is done in the transformation
stage. After this the data is loaded into the DW.
process has to be carried out table wise. The dimensions must be loaded first. Only then the facts can be loaded. Each surrogate
key in the fact table must a corresponding key in the dimension table. In other words the fact table has a foreign key relationship
with a Dimension table. The keys in the fact table are loaded from the corresponding dimension tables.
The ETL process
can be a 2-stage process. In the 1st stage, an intermediary set of tables called staging tables are loaded. In
the 2nd stage the dimensions and facts are loaded. The staging tables contain cleansed data.