I've created a shell dimension with a nonaggregatable attribute and SCOPEd each of the members. What we've been finding is that as long as the user only uses 5 or fewer members in a single query everything works great - at most just a couple seconds of runtime.
However, at *6* members or greater we immediate get the error:
"The expression contains a function that cannot operate on a set with more than 4,294,967,296 tuples"'
I've already went through my SCOPEs as much as possible, there's not really much else I can do and obviously it works quite well until I hit the limit in the engine (which I suspect is a bit arbitrary since 5 runs great). So this leads to two questions:
1) Does SSAS 2012 still have this same limitation?
2) My cube currently has 10 million facts but performance is not an issue so I have not added the complexity of partitioning yet. Any idea if partitioning would "parallelize" this execution? I would partition on year and all SCOPE statements are SCOPEd to a particular year.
Thanks,
Jusin