We have several dimensions in our OLAP database.

For most of the dimensions, the updating time is realtively quick. However,
for others it's slow since it reprocesses all the underlying partitions.

Time is broken down like this:

--Processing Dimension "Time" completed successfully -duration 20 seconds
--Processing Cube "Sales" completed successfully --duration 2 seconds
--Processing Measure Goup "Sales measures" completed
uccessfully --duration 1 second


However other dimensions look like this:
--Processing Dimension "Business Unit" completed successfully -duration 7
seconds
--Processing Cube "Business Unit" completed successfully --duration 28
minutes
--Processing Measure Goup "Sales measures" completed
uccessfully --duration 27 minutes
--Processing Partition 'D_20060504' completed successfully.
<and so forth for every partition>


My question is...why do some dimensions require the processing of the
individual partitions while others do not? The processing of the other
dimensions is killing our load time.

This is something I need to understand asap since we're planning on rolling
it into production.

Perhaps some dimensions have aggregations defined on them but others do not?
When a dimension is updated, aggregations/indexes may need to be rebuilt
depending on the changes to the dimension.

Thanks,
Akshai
--
Try out the MSDN Forums for Analysis Services at:
http://forums.microsoft.com/MSDN/Sho...ID=83&SiteID=1

This posting is provided "AS IS" with no warranties, and confers no rights
Please do not send email directly to this alias. This alias is for newsgroup
purposes only.

"Jesse O." <jesperzz (AT) hotmail (DOT) com> wrote

That was my first thought, however that's not the case.

The only thing I can think of is the two dimensions that do require all the
partitions to be processed is that they are both part of a reference
dimension. Dimension "Business Unit" is a reference dimension of dimension
"Sales". Sales is not a reference dimension off of anything else.

Do reference dimensions or a dimension that is part of a reference dimension
required all the partitions to be processed after the dimension is updated?





"Akshai Mirchandani [MS]" <akshaim (AT) online (DOT) microsoft.com> wrote

I think if they are materialized (i.e. not Indirect), then yes. The
relationship between the referenced dimensions could have changed and so the
data in the partitions may not match anymore without re-processing.

Thanks,
Akshai
--
Try out the MSDN Forums for Analysis Services at:
http://forums.microsoft.com/MSDN/Sho...ID=83&SiteID=1

This posting is provided "AS IS" with no warranties, and confers no rights
Please do not send email directly to this alias. This alias is for newsgroup
purposes only.

"Jesse O." <jesperzz (AT) hotmail (DOT) com> wrote