Hi,
A transactional cube is loaded data from BPS (APO* request)all the time, and there are aggregates associated with this cube. By default aggregates are rolled-up after the number of records reaches 50000. If a query is run before that threshold is reached, is it going to hit the aggregates? If yes, it will not have the most recent data; if no,performance would be very slow...why do we need the aggregates since they are not accessed?
Am I missing something? Could you guys clarify.
Thanks,
FF