TM1 Forum

Hello All,

Can we fine tune the performance of Filter by Attribute ?

Currently I have a dimension which has 1.4 millions elements in a dimension with 15 Attributes. When i am trying to filter by attribute in Subset Editor it is taking around 4 minutes to come up with popup box with list of attributes. is there a way we can fine tune the response time so that it pops up back really fast?


Regards,
Firefly

I'd just have a TI that runs as regularly as attributes get updated which cycles through and creates static subsets for each possible attribute selection. If that's a route you go down just be sure to combine the name of the attribute column and contents so that you aren't hindered by the possibility of having the same selection in different types of attributes (which is highly likely with flag attributes etc.)

Subset Editor performance on large dimensions is a bit of a known annoyance. There's no config setting or tweak, as far as I know, that will speed it up for you. However, you could build hierarchies into your dimension based on the attributes such that all elements with a particular attribute are children of a consolidation. Then, instead of searching the dimension each time you want those elements you simply select the children of that consolidation. Have you tried that approach?

Thank you all for your suggestion i will be perusing the route of alternate Hierarchy as suggested by rmackenzie.