I'm a new user of analysis server and I'm getting stumped on something that I'm hoping some more experienced users can help me out on.
My basic situation is that I've got two dimensions, each with 4 levels, that I need the aggregation method to be an average. I can get this to work using the custom rollup formula, avg(<dim>.CurrentMember.Children). However, from what I can gather the preaggregation when using a custom rollup only occurs within a dimension, not across dimensions. When I browse the data after defining the average aggregation for both dimensions the queries are way too slow. In fact all the clients I have tried effectively freeze at this point taking up 100% of the CPU.
Is there anything I'm missing here? A way to get across dimension precalculated aggregations? Another (better) way to implement an averaging aggregation?