I have recently begun a new project where I am working on a SQL Server Analysis Services (SSAS) implementation.
I was trying to use the aggregation designer to...well...design some aggregations but I was getting an error:
Errors in the metadata manager. The database with the name of 'foo' already exists in the 'bar' server.
That seemed a little strange to me - why should the presence of an already-deployed database affect the aggregation designer within a new cube that I was building? Well, the answer was provided to me by Edward Melomed and you can read it here.
The interesting thing to note here is that the algorithm for designing aggregations utilises the already-deployed database. Now as far as I am aware (my SSAS knowledge is a little rusty) that is different behaviour from SSAS2000 so its something to be aware of.
More details as and when I uncover them...
-Jamie