Page 15 - DCAP603_DATAWARE_HOUSING_AND_DATAMINING

Unit 1: Data Warehouse Practice

would bring down the scope of the project to something smaller and manageable, yet be scalable notes
to gradually upgrade to build a comprehensive data warehouse environment finally?
The recent trend is to build data marts before a real large data warehouse is built. People want
something smaller, so as to get manageable results before proceeding to a real data warehouse.
Ralph Kimball identified a nine-step method as follows:

Step 1: Choose the subject matter (one at a time)
Step 2: Decide what the fact table represents
Step 3: Identify and conform the dimensions
Step 4: Choose the facts

Step 5: Store pre-calculations in the fact table
Step 6: Define the dimensions and tables
Step 7: Decide the duration of the database and periodicity of updation
Step 8: Track slowly the changing dimensions
Step 9: Decide the query priorities and the query modes.

All the above steps are required before the data warehouse is implemented. The final step or
step 10 is to implement a simple data warehouse or a data mart. The approach should be ‘from
simpler to complex’,
First, only a few data marts are identifies, designed and implemented. A data warehouse then
will emerge gradually.
Let us discuss the above mentioned steps in detail. Interaction with the users is essential for
obtaining answers to many of the above questions. The users to be interviewed include top
management, middle management, executives as also operational users, in addition to sales-
force and marketing teams. A clear picture emerges from the entire project on data warehousing
as to what are their problems and how they can possibly be solved with the help of the data
warehousing.
The priorities of the business issues can also be found. Similarly, interviewing the DBAs in the
organization will also give a clear picture as what are the data sources with clean data, valid data
and consistent data with assured flow for several years.

Task Discuss various factors play vital role to design a good data warehouse.

1.6 Data Warehouse architecture

1.6.1 Why do Business analysts need Data Warehouse?

A data warehouse is a repository of an organization’s electronically stored data. Data warehouses
are designed to facilitate reporting and analysis. It provides many advantages to business analysts
as follows:
1. A data warehouse may provide a competitive advantage by presenting relevant information
from which to measure performance and make critical adjustments in order to help win
over competitors.

LoveLy professionaL university 9