Hi,

I'm new to OLAP and Cubes, etc

Currently I am working on a project for an existing SQL database, that
has a few poor design flaws to it.

The organization has a hierarchical arrangement of Pillars -> Lines of
Business -> Departments -> Products. Each tier is stored in a separate
table and the schema has similarities and differences.

How do I create a hierarchical dimension in MS SQL Analysis Manager
(i.e. What type of Schema to use) whereby I can query Products or roll
up the aggregates and at each level display the corresponding
information?

Do I need to create any additional tables since this DB is not
originally meant for OLAP?

Thanks,

Lawrence

You have already mentioed that there few poor design flows so adding
new tables may increase the complexity and create other issues in the
future. There are many factors we need to take into considerations
including the budget.

Before creating an OLAP there lots of background work we need to do in
order to decide how to go forward.

First identify the Subject Models (for example what are the Subjects
we need to include such as Sales, Marketing, HR, etc)
Develop Business Model (Add time, demormalization)
Create System Model (what apporach should use MOLAP, ROLAP, etc)
Create Technology Model (what technology to use)

Do I need to create any additional tables since this DB is not
originally meant for OLAP?
If the DB is not originally meant for OLAP but OLTP, the possible
issues going to be performance. In OLTP we see all normalized
transaction oriented tables and indexes etc, but in OLAP the focus
should be Decision making so reporting and retrieving for a particular
subject area is very immportant. In OLAP de-normalized dimensional
approach is ideal which will be helpful to present data of interest to
particular subject area.

Regards
~ PV


virtualawrence (AT) gmail (DOT) com (Lawrence) wrote in message news:<64cf894d.0412010750.5469e8e1 (AT) posting (DOT) google.com>...