hi experts,
In my cube when i do the full process it is taking too
much time.when i do incremental update it is adding double
the data to the existing cubes data.(without adding the
data to the fact table).But is there any approach or
options through AS,when i do incremental the data remains
same(same as fact tables data).How far the cube partitions
will work on this?pls give your valuable suggestions.

When you incrementally process your cube, you expose ONLY THOSE FACTS which
you want to add. Since you are exposing the entire table, everything is
being double counted. Typically I recommend one of two techiques: 1) have a
dedicated table which contains just the rows you want to add during the
incremental process (and then delete them afterwards) -- remember that you
also need to add them to the fact table since they need to be there if you
ever need to do a full process; or 2) that DBAs use a view to control what
data is exposed. For the view technique, you might have a control table in
the RDBMS which has beginning date/time values, and join with the fact table
to expose only those records after a particular date/time. When you want to
do an incremental process, you update the control record in the RDBMS and
then do your incrementals. If you want to do a full process, you use the
fact table directly.

Hope this helps.
--
Dave Wickert [MSFT]
dwickert (AT) online (DOT) microsoft.com
Program Manager
BI SystemsTeam
SQL BI Product Unit (Analysis Services)
--
This posting is provided "AS IS" with no warranties, and confers no rights.

"Robert" <anonymous (AT) discussions (DOT) microsoft.com> wrote