Hi there
1-If I have 300 expected users for an application runnning on MS Analysis
services OLAP with each user having a different profile as to what data and
cubes he can access. The whole application logs in through a sinlge windows
account user. In such a case, would it make sense for me to create a virtual
cube for each user as per his profile. I mean, is this a feasible solution
to create so many virtual cubes. If not, are their any alternate solutions.

2- While using Analysis services, is there any way for me to identify the
level (on each dimension) to which the aggregations can be made. Or is it,
that i can only define aggregations level in terms of percentage and size of
the data.

3- I want to create a repository wherein I will keep the profile of all
expected users as to what data (within a cube) and cubes they they access.
What is the best way for me to populate this repository directly from
MSOLAP. I mean I want to replicate the dimensional structure of each cube in
the repository - how best and easily I can do so on a continual and
automated basis.