May 22, 2009 at 9:50 am
I'm new to DW, I'm reading kimball and the concepts dont seem overwelming (yet) but I get the feeling that "Data Mart", "cube" and OLAP are 3 different things and they have precise definitions that I'm not catching. Immon even draws them as separate things.
Once you're done laughing at the newbie; can you enlighten me? :))
tks.
July 8, 2009 at 10:29 am
to extract this to a 'Noddy' style answer.
Data Mart is a database
Cube is a method of storing OLAP data. You could call it a OLAP table.
OLAP means anlytical rather than transactional data and you are looking at the data stored in a cube.
In a not so noddy way
Data Mart is a collection of data that may come from many sources and is a little data warehouse in its own right. So there may be all the data relating to sales in one data mart.
From this data mart you would identify the fact data (sales) and the dimension data (shops, sales dates, sales persons, etc). Then you bild a cube that contains the facts and dimensions. The cube actually physically contains the data and becomes what you could call a multidimensional table. You can query this by each of your dimensions to reveal the facts.
OLAP generally refers to the type of system you are building. In the same way that RDBMS tells you that you are dealing with a relational database OLAP tells you that you are dealing with a multidimensional database.
Hope that helps.
Ells.
😎
July 9, 2009 at 9:21 am
very concise. Very helpful. tks much.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply