Hi,

Can someone please explain the 'Staging Area' concept. Currently I
populate my Data Mart by running a DTS which performs many tasks and at
the end of which the fact table is updated and the cube is processed...

However reading a post/article somewhere I get the feeling that I should
use the Staging Area concept/approach. Can someone kindly explain this
with maybe a brief example and additionaly a link to a RELEVANT article
one the web.

Many Thanks.

Hi,

in a few words, staging area is all about preparing data. You could say that
staging area is all about your work with ETL tool.

For example how do I use staging area:
My data warehouse, I call it CDL (common data layer) consists of 3
databases:
Staging - for preparing data
App - for all applications needed with your data warehouse, analytical
solution
DW - clean/pure data for your analytical solution - OLAP

Now all your work with transfering data from different sources and cleaning
and transforming of that data is done in staging.
At that point when you have your data prepared you transfer it to the DW
database. And from there on you are building your OLAP cubes and data marts.

Hope it helps.

Peace,
Andrej




"Learner" <wantnospam (AT) email (DOT) com> wrote

The bit about the staging area (relationel base) and the DW area (also a relationel base) i fully aggre

But the App I'm not sure I get....if it's the data suppliers typically these applications would be placed on other servers ... but perhaps this is a database for tools which require an sql base ? What would that be ? If I have some tools I would just install them on the DW-server

Lots of things can be done in the staging area regarding cleansing but one of the things that differentiate subject from staging is among other things the surrogate keys ... in a dimension table it is considered best practice to add a "dummy" key and place this "dummy" key in the fact table and use this key to combine the fact and dimension table.

\Michael Vardinghus

For sure one of the methods that should be done is to implement surrogate
keys at staging area.
But I have to say that this is something that you should think about if you
need it.
My experiance is that in small cubes this is just waste of time. In
enterprise solutions this is a must.

Yeah, about the app. As I was talking about it, I meant all about staging.
But the existance of App is a part of CDL (Common Data Layer). This
represents a database for small applications that are developed as part of
BI solution. With each BI solution we are making we get some requests that
can not be made in OLAP, but can be made with some application that uses
data from transactional system and OLAP system. That is why I mentioned it.

Peace,
Andrej


"Michael Vardinghus" <anonymous (AT) discussions (DOT) microsoft.com> wrote

a database for tools which require an sql base ? What would that be ? If I
have some tools I would just install them on the DW-server.
the surrogate keys ... in a dimension table it is considered best practice
to add a "dummy" key and place this "dummy" key in the fact table and use
this key to combine the fact and dimension table.