Building Basic Dimensions and Cubes - Expert Cube Development with SSAS Multidimensional Models - page 58

Database processing
Assuming you've set the Processing Option project property to Do Not Process ,
then the result of deployment will be a new Analysis Services database on the server
containing a cube and several dimensions. You won't be able to browse the cube yet
though, as all that has been created are empty structures which need to be loaded
with data from your data warehouse. In order to load the data, you have to process
the objects and you can do this easily from within SSDT by going to the Database
menu and clicking on Process . Once you've done this, the Process Database dialog
appears. Leave the Process Options column showing Process Full , click on Run , and
everything in the database should be processed.
As an Analysis Services database is nothing more than a collection of objects,
processing a database involves nothing more than processing all of the cubes and
dimensions in the database. In turn, cubes are made up of measure groups, which
are made up of partitions, and dimensions are made up of attributes, and all of
these objects have their own discrete processing operations. As a result, processing
a database can kick off a lot of individual processing jobs and by default, Analysis
Services will try to do a lot of this processing in parallel to reduce the overall time
taken. A Process Full will always drop any data currently in an object and reload
its data from scratch. It's the most time-consuming form of processing and there are
a lot of other options here that we can use to reduce processing time, which we'll
discuss in Chapter 10 , Going in Production .
You can watch all of these processing jobs executing in the Process Progress window
that appears after you click on Run in the Process Database dialog. You'll see each
object that is being processed listed, as well as the time processing started and ended.
If you expand each object, you'll eventually find the SQL queries run to retrieve
data from the data warehouse. If you select any node and click on the View Details
button, you'll see a new window appear, containing all of the text for the node, and
this is very important when it comes to viewing long SQL queries or messages, as
shown in the following screenshot.
As you can't do anything else in BIDS while processing is taking place, it can be a
good idea to just do a Deploy from there and then start processing separately in
SQL Management Studio. This way, you can carry on developing while processing
is taking place.