Hi,
I'm facing a little feeder issue. Lets say that Cube A is a very small cube to indicate a distribution percentage, with only two dimensions. In my second cube, I want to use this percentage and apply it to all elements on a new dimension. Creating the rule is no problem, writing the feeder from the first cube is, since not all dimensions for the second cube are recognized. For example:
Cube A (second dimension is a measures dimensions with one element 'Percentage'):
Total 100 %
Belgium 20 %
France 40%
UK 30%
Other 10%
Cube B contains a measures dimension with an element where i would like to see the percentage from the first cube. But this cube also contains for example a projects dimensions with +100 items. So the rule in Cube B would be: ['Percentage'] = DB('Cube A', !Countries, 'Percentage');. Because project is not explicitly defined, TM1 would link in the same percentage for all projects.
But the feeder in Cube A would then have to be: ['Percentage'] => DB('Cube B', !Projects, !Countries, 'Value');, which returns a syntax error because the Projects dimension is not recognized. Is there a workaround for this so that I don't have to write a feeder for all +100 projects individually?
Thanks for the help,
Mathias.
Feeders between cubes without the same dimensions
-
- Posts: 27
- Joined: Sat Oct 02, 2010 3:05 pm
- OLAP Product: Cognos TM1
- Version: 10.1
- Excel Version: 2010
-
- Site Admin
- Posts: 1454
- Joined: Wed May 28, 2008 9:09 am
Re: Feeders between cubes without the same dimensions
What are you doing with the percentage? If you are applying it to a value which is input or fed, you don't need to feed the percentage as well.
-
- MVP
- Posts: 2832
- Joined: Tue Feb 16, 2010 2:39 pm
- OLAP Product: TM1, Palo
- Version: Beginning of time thru 10.2
- Excel Version: 2003-2007-2010-2013
- Location: Atlanta, GA
- Contact:
Re: Feeders between cubes without the same dimensions
Just like with any intercube rule, if both cubes don't have exact same dimensionality, you have to specify an exact element for the unmatching dimensions. In your case you will have to specifically list a member(s) of the
Projects dimension. To make it easier on you in these types of situations, TM1 allows you to point to a consolidated node in the unmatching dimensions which then will feed all the children of that node. So, if you don't already have it, add a node to the Projects dimension called "All Projects" which has all the projects as it's children. Then your feeder statement would look like:
['Percentage'] => DB('Cube B', 'All Projects', !Countries, 'Value');,
Projects dimension. To make it easier on you in these types of situations, TM1 allows you to point to a consolidated node in the unmatching dimensions which then will feed all the children of that node. So, if you don't already have it, add a node to the Projects dimension called "All Projects" which has all the projects as it's children. Then your feeder statement would look like:
['Percentage'] => DB('Cube B', 'All Projects', !Countries, 'Value');,
-
- Posts: 27
- Joined: Sat Oct 02, 2010 3:05 pm
- OLAP Product: Cognos TM1
- Version: 10.1
- Excel Version: 2010
Re: Feeders between cubes without the same dimensions
Thanks! I tried the 'All Projects' solution, worked like a charm!
-
- Regular Participant
- Posts: 177
- Joined: Thu Nov 05, 2009 8:45 am
- OLAP Product: IBM COGNOS TM1
- Version: 9.5
- Excel Version: 2007
Re: Feeders between cubes without the same dimensions
Dear Mathias,
Please can you explain me in detail. I am also trying on this.
Thanks for your help.
Thanks and regards
TM1 9.5
Windows XP
excel 2007.
Please can you explain me in detail. I am also trying on this.
Thanks for your help.
Thanks and regards
TM1 9.5
Windows XP
excel 2007.
-
- Posts: 27
- Joined: Sat Oct 02, 2010 3:05 pm
- OLAP Product: Cognos TM1
- Version: 10.1
- Excel Version: 2010
Re: Feeders between cubes without the same dimensions
As tomok already mentioned, when you create a rule between two cubes that don't share the exact same dimensions, you need to specify a single element of the missing dimensions in order for the rule to work. The same applies for the feeders, but in the opposite direction.
In my case, I have two cubes with a seperate set of dimensions:
Cube A contains percentages per country based on number of employees for example. I wanted to use this percentage in a second cube, where I also have the country dimension but also a projects dimension. The purpose is to use the percentage as a parameter to distribute certain costs per project along the different countries. So per project I could easily say that in cube B: ['Percentage'] = N: DB('Cube A', !Countries, 'Percentage'). This would take the percentages from Cube A and use the same distribution for every project. No problem here.
But this rule should also be fed from Cube A. The problem there is that Cube A doesn't contain the Projects dimension so writing a feeder like ['Percentage'] => DB('Cube B', !Projects, !Countries, 'Percentage') would not work because Cube A doesn't recognize the Projects dimension. Luckily you do not need to write a feeder for every project individually (e.g. ['Percentage'] => DB('Cube B', 'Project 1', !Countries, 'Percentage'), DB('Cube B', 'Project 2', !Countries, 'Percentage'), ...). If you have a top element in the projects dimension (i.e. 'All Projects') you can write the feeder to target this element, which then would mean that all underlying elements are also fed. So the only feeder you'll need is then ['Percentage'] => DB('Cube B', 'All Projects', !Countries, 'Percentage').
Regards,
Mathias.
In my case, I have two cubes with a seperate set of dimensions:
- Cube A: Countries, Measures.
- Cube B: Projects, Countries, Measures
Cube A contains percentages per country based on number of employees for example. I wanted to use this percentage in a second cube, where I also have the country dimension but also a projects dimension. The purpose is to use the percentage as a parameter to distribute certain costs per project along the different countries. So per project I could easily say that in cube B: ['Percentage'] = N: DB('Cube A', !Countries, 'Percentage'). This would take the percentages from Cube A and use the same distribution for every project. No problem here.
But this rule should also be fed from Cube A. The problem there is that Cube A doesn't contain the Projects dimension so writing a feeder like ['Percentage'] => DB('Cube B', !Projects, !Countries, 'Percentage') would not work because Cube A doesn't recognize the Projects dimension. Luckily you do not need to write a feeder for every project individually (e.g. ['Percentage'] => DB('Cube B', 'Project 1', !Countries, 'Percentage'), DB('Cube B', 'Project 2', !Countries, 'Percentage'), ...). If you have a top element in the projects dimension (i.e. 'All Projects') you can write the feeder to target this element, which then would mean that all underlying elements are also fed. So the only feeder you'll need is then ['Percentage'] => DB('Cube B', 'All Projects', !Countries, 'Percentage').
Regards,
Mathias.
-
- Site Admin
- Posts: 1454
- Joined: Wed May 28, 2008 9:09 am
Re: Feeders between cubes without the same dimensions
I'm going to be a bit pushy here
What are you doing with your percentage? If you are using it in (say) an allocation calculation, and you are feeding the cell you are allocat_ing, you don't need to feed the percentage as well. Percentages are typically poor feeder sources as they they are less sparse than the values. All together now....
"Only ONE value in a calculation needs to be fed"
You may have a small system where this doesn't matter. If so good luck, but sometime you will encounter one where it matters a lot.
What are you doing with your percentage? If you are using it in (say) an allocation calculation, and you are feeding the cell you are allocat_ing, you don't need to feed the percentage as well. Percentages are typically poor feeder sources as they they are less sparse than the values. All together now....
"Only ONE value in a calculation needs to be fed"
You may have a small system where this doesn't matter. If so good luck, but sometime you will encounter one where it matters a lot.
-
- Regular Participant
- Posts: 177
- Joined: Thu Nov 05, 2009 8:45 am
- OLAP Product: IBM COGNOS TM1
- Version: 9.5
- Excel Version: 2007
Re: Feeders between cubes without the same dimensions
Dear Mathias,
thank you very much.
thank you very much.