I read somewhere that you need to idenfity a dimension as a measure in TM1 cube before publishing it to cognos 8. It didnt mentione why or what would happen if we dont do so. Could anyone suggest the reason for doing so
If you don't do it, the query will fail - good enough reason? :)
Why it's required I don't know for sure, but my guess is that MS Analysis Services cubes always have a measure so Cognos' MDX implementation assumed one exists.
I do recall a recent tm1s.cfg parameter which automatically sets the last dimension to be the measure, but really, it's not a tough thing to do.