Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

usage based aggregation optimisation (SSAS 2005)

1 view
Skip to first unread message

Antoon

unread,
Mar 16, 2010, 5:20:01 AM3/16/10
to
To minimise the impact for users we always process our cubes on server A and
then backup/restore them to server B.
Users only access server B.

Now I wan't to use "usage based aggregation optimisation". The user
statistics are obviously gatherd from the cube on server B. But I would like
to use these to optimise the cube on server A.

I can link the cube on server A to the same statistics tables of the cube on
server B. But I notice that in de logging table the name of the server is
included in the objectpath. Do I presume that the optimisation of cube A will
only use the statistics of cube A.

Is it possible to use the statistics of Cube B to optimese Cube A?

Regards

Hight

unread,
Mar 17, 2010, 5:06:34 PM3/17/10
to

Have you tried setting the Log \ QueryLog \ QueryLogConnectionString
to the same location as Server A. If you set the usage statistics
location to be the same on both servers then it should use them on
Server A when creating aggregations and still be logged to by Server B.

amish

unread,
Mar 22, 2010, 5:53:46 AM3/22/10
to

In server property Log\QueryLog\ConnectionString you can specify the
server detail of server B where you want to log query.

Amish Shah
http://shahamishm.blogspot.com


0 new messages