Hi Team,
we are facing attached error while building cube manually in production, can you please help me with the fix.
TR0417) The supported limit of categories in the model has been exceeded (2 million)
Quote from: krishdw85 on 12 Jul 2017 11:54:01 AM
Hi Team,
we are facing attached error while building cube manually in production, can you please help me with the fix.
TR0417) The supported limit of categories in the model has been exceeded (2 million)
http://www-01.ibm.com/support/docview.wss?uid=swg21652531
I am new to model design or fixes, can you please explain step by step and which one would be helpful. This cube is using from many users and it is very critical one. Thanks for your help in adv.
•Reduce the category count within the model by removing unused dimensions, levels or categories
•Remove lowest detail levels and use drill-through instead
•If the model is responsible for several cubes with differing scope, dimensions and measures, consider splitting the model such that it is specific to a single cube
•Place filters on the data source to restrict the incoming categories
Quote from: krishdw85 on 14 Jul 2017 10:19:14 AM
I am new to model design or fixes, can you please explain step by step and which one would be helpful. This cube is using from many users and it is very critical one. Thanks for your help in adv.
•Reduce the category count within the model by removing unused dimensions, levels or categories
•Remove lowest detail levels and use drill-through instead
•If the model is responsible for several cubes with differing scope, dimensions and measures, consider splitting the model such that it is specific to a single cube
•Place filters on the data source to restrict the incoming categories
Hi,
That's worrying. If you're new to model design, why are you working on a cube that's in production (ie live)? Isn't that hugely dangerous?
The error you're reporting means you have hit the physical limit for categories (dimensional members) within the dimensions of your Transformer model. The solutions listed above are simply ways of making you think about how you can reduce the number of categories. There is no step-by-step answer per se - you need to understand what the requirements for the cube are and tally those with the physical limitations of your model and cube to be able to decide which of these options (if any) is the right one for you. Do you have any unused dimensions/levels/categories in your model? Do all the dimensions need all the levels you are currently including? Does your model feed multiple cubes? Do you need the entire range of data currently being collected (eg what time periods are required and are there any years currently in the model that you don't need)?
Nobody but you can answer these questions...
MF.