Here’s a complete list of all the file extentions which SSAS uses. You may want to exclude these when you’re running an anti-virus scanner on your server:
A list of all the ‘Data Directories‘ can be found here:
Hi there reader!
There are a couple of methods on how to populate so called ‘ragged hierarchies’ in SSAS, a few tricks:
Naturalizing a Parent-Child Hierarchy:
TopDown and hide middle levels:
BottomUp approach by Chris Webb (Ragged Hierarchies, HideMemberIf and MDX Compatibility):
I really liked the BottomUp approach that Chris Webb proposed. This is how a TopDown-hierarchy looks like:
And here’s the BottomUp-hierarchy we’re going to be using for this demonstration:
First things first, here’s the cube’s structure:
Define the relationships (nothing fancy here):
Here’s where the magic happens! Set the HideMemberIf-property to ‘OnlyChildWithParentName‘, so basically hide the child when it has the same name as its parent.
And here’s the result of that hard work:
This method works for Excel without setting the “MDX Compatibility = 2” requirement, but in SSRS and your MDX query it will only return those records which have no hidden levels. Strange issue, I suggest you make another seperate hierarchy for reporting that has the HideMemberIf property set to ‘Never’ for all levels.
I’ve Googled for MDX Cheat Sheets, found a couple, but never exactly what I was looking for.
So I made my own, hope this one explains things a lot easier for you (and also as a reference for myself) 😉
For the demo, I’m using a Year(2014)/Month(201401) hierarchy in my SSAS cube which looks like:
Here’s a more graphical explanation on how things work:
SSAS MDX Cheat Sheet
Here’s another MDX Cheat Sheet I posted after, all about Calculated Measures:
Thanks Denny Glee for sharing this.
It’s old news, but still it’s very cool 🙂
I always thought that Yahoo! had a 7TB SSAS Cube, but it has been growing of course.
As of dec 2012, it reached 24TB!!
Microsoft Case Study:
Yahoo! Improves Campaign Effectiveness, Boosts Ad Revenue with Big Data Solution
Some key numbers from this case study include:
24TB Analysis Services MOLAP cube
2PB source data of a 14PB Hadoop cluster
700M unique users, 47% of the global online population
3.5B ad impressions/day
More slides can be found here.