Day 1: 5:49pm – Progress, and then 28 Errors and 7 Warnings
Spent the morning and afternoon reading up on the basic elements of a multidimensional model, including the data warehouse, facts, measures, dimensions, attributes, hierarchies, cubes, etc. The lingua franca of SQL Server Analysis Services is a bit of a headache, but I’m sure I’ll get used to thinking in those terms before too much longer. Two-dimensional life in OLTP was so much simpler: rows and columns!
I ‘modeled’ my first schema on a notepad at the pub across the street from the hotel, while taking down a cheeseburger. Although it looks more like a grocery list than a star or snowflake schema, it’s a start. From what I’ve read, I don’t understand why everyone wouldn’t want a snowflake schema, to take advantage of more complex hierarchies, but I’m in no position to be asking that question yet.
Poor Man’s Schema
Once back at the hotel, I created a data warehouse in the relational database engine in SSMS from a copy of our production OLTP database. I’m not worrying about using SSIS to load the data warehouse from the OLTP source yet – I’ll save that for later.
From there, I jumped over to SSDT, and in about 30 minutes, quickly knocked out the following:
- Created an Analysis Services Multidimensional Project
- Created a data source
- Created a data source view
- Created a cube
- Created some dimensions
All of the wizards in SSDT made it very easy, and I had a ‘this is too good to be true’ feeling, but nonetheless couldn’t resist feeling a bit pleased with myself for having filled the design surface with so many shiny new things.
Progress!
My fears were confirmed when I went to deploy the solution back to the server and received no less than 28 errors and 7 warnings:
Oh dear.
That’s a lot of errors and warnings. So, I’m going to take a break, find some food, and dive back in later tonight when my confidence has recovered.
OLAP Sprint 