Posts Tagged SSAS

Analysis Services : How to break distinct count measures into separate measure groups ?

Having blogged last week here about the advantages of the new storage file format found in the Denali’s CTP3 for those kind of measures when they are of the string data type, remember me that maybe this little feature may need a post. This is no rocket science, just a little tip that every one would solve after spending 5 minutes on it. I even think that most of you may already know about it….

Read the rest of this entry »


Comments (2)

[SQLServer 2012] Analysis Services New Feature : Scalable String Storage

As you may know, both dimension’s attributes and distinct count measures can be of string data type. During processing, Analysis Services reads data and put them into individual stores organized by attributes in case of dimensions or by measures groups/partitions for cubes.

The file format of those stores differ to adapt to the type of the data located inside. By default, until now, string stores couldn’t exceed the 4GB file size limit.

Starting with the CTP3, a new feature names “Scalable String Storage” allow to go beyond this limit. A new property named “StringStoresCompatibilityLevel” is available at different places to activate it.

I’ll review in this post how to set up this new feature in your SSAS solution and what we can expect from it…

Read the rest of this entry »

Comments (5)

Analysis Services Feature Suggestion : Managing Measure Group visibility

The most commonly used reporting tool by the users to query Analysis Services may surely be Excel. As professionals, while testing SSAS solutions, we also add BIDS and SSMS to the list.

One of the great but little known feature of those SSAS reporting tool, is their ability to hide the multiple measure groups complexity by selecting one of them and only displaying the measures and dimensions that belongs to it.

Just in case you don’t know what I’m talking about, here’s a screenshot :

Read the rest of this entry »

Comments (1)

Analysis Services: What are the best practices for grouping measures ?

You’ll find in the following post some general considerations to have in mind when designing cubes about how to dispatch indicators into groups to suite the end-user needs and the consequences of these choices.

Lately, a colleague of mine at Avanade was embarrassed because his customer wanted to visually group and mix heterogeneous kind of indicators into same folders so that he’ll later find them easily into his pivot table into Excel.

As a good practices follower he designed his data warehouse and cube by respecting multidimensional modeling, commonly approved, best practices and putted all the measures that share the same grain into a single fact table and into the same measure group…. which, as he discovered it later, is not necessarily the organization that the end-user have in mind for his data when it comes to build analysis reports.

My colleague first thought was to rely on the “display folder” and “associated measure group” properties to fill the end-user need. The drawback is that what is defined in a measure group stays in the measure group and a fact table cannot lead to more than one measure group.

So as it comes that he was unable to easily answer the requirements, he asked me for advises on how to handle the situation. Here is what we can say…

Read the rest of this entry »

Comments (2)