I have an OLAP Cude that takes about 1 hour and 10 minutes to do a full
process. Now while that isn't that bad, what I find odd is the log file from
the DTS package that runs the full process, it takes about 10 minutes to
process all the partitions, but there is an hour long wait or pause in the
log until it commits the transactions.

Here is a sample of the log:

Error = 1 (00000001), Description = 5/31/2005 9:49:29 PM 13882
Completed Processing Cube 'Sales_Analysis'. End time: 9:49:29 PM Duration:
0:10:12
Error = 1 (00000001), Description = 5/31/2005 10:49:29 PM 13882
Committing transaction in Database 'Sales_Analysis'
Error = 1 (00000001), Description = 5/31/2005 10:49:44 PM 13882
Committed transaction in Database 'Sales_Analysis'

What I find odd is the exactness of the deal, 1 hour to the second.

Has anyone seen something like this or have an idea as to what is happening.

I'm running SQL Server 2000 with SP3.

TIA

Do you have many aggregations on the cube?


"Michael" <michaelteff (AT) skyline (DOT) com> wrote

I'm not 100% up on OLAP and AS, so I will try to best answer the question.

Yes I believe there are, in the cube editor there are about 20 dimensions,
18 measures(aggregations) and 12 calculated members.

And I do realize that these can have an effect on performance, I'm currently
reading through the guide on Analysis Services Performance to do some tuning
and performance monitoring.

I just thought it was very odd that the log showed an exact 60 minute, to
the second, every time, gap between processing and committing.

Thanks.

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