We have deployed MS AS version 8.00 cubes to our users. We are not
very impressed with the performance. Sometimes it takes 8 seconds for
a view to come up. The cube has 6 dims and 4 measures. We have tried
all 'aggregations option' i.e. processed the cube for 'performance
gain reaches maximum' option. but didn't see any performace change.
Also MS AS cube builder crashes frequently, as a result we have to
bounce the server quite often.
Could somebody sugegst ways to troubleshoot this problem. At this
point we are ready to evaluate anything --cube design, server
configuartion or any other factor..
Is anybody aware of any guidelines or tips?

Can you talk a bit more about what you are seeing?
Is it with querying or processing? Is it on the server itself or just
remotely? Is it when browsing with Analysis Manager or when using a
front-end tool?
--
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.

"Rinki Jain" <rinkijain (AT) gmail (DOT) com> wrote

Thanks for responding.
Here's what happening
During cube developement the client works slow, and AS sometimes hangs
while processign the cubes. We haev to 'end the task' on client and
start it again.
But this is not a regular phenomenon but happens quite often.

Secondly , the performance of the cubes at run time is not very great.
If I am not wrong, when '100% performance gain' option is used while
designing the storage ,it should pre calculate everything in the cube
and should be able to present quickly.

What other things can we do to speed up the cubes at run time?
Rinki Jain
Project Manager, Decision Support

"Dave Wickert [MSFT]" <dwickert (AT) online (DOT) microsoft.com> wrote

All of this is discussed in detail in the Analysis Services Performance
Guide, see:
http://www.microsoft.com/technet/pro.../ansvcspg.mspx

There is also some good material around how to configure your system in the
Operations Guide
http://www.microsoft.com/technet/pro.../anservog.mspx

Some topics to look into:
1) setting up the system-wide processing log file so you can keep track of
long-term history of processing performance.
2) using the optimize schema tool (from the cube editor) to simplify the SQL
statement used for processing.
3) memory utilization -- which could be causing the hangs
4) make sure you are running SP3 (the operations guide tells you how to
check this) -- there is a separate download for Analysis Services; just
because the RDBMS is at SP3 doesn't mean that Analysis Services is.

The 100% performance gain will NOT calculate all of the theoretically
possible aggregates. It designs 100% of the *effective* aggregates. See the
performance guide for a discussion of the 1/3 rule. In general, we recommend
that you start out at 20% and only increase it if you can demonstrate actual
increases in client usage. You get your most bang for the buck around 20%.

--
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.

"Rinki Jain" <rinkijain (AT) gmail (DOT) com> wrote

Dave
We are a Nashville, TN based hospital and university. We would like to
have a consultant come on site and look at the AS configuration and
resolve the issue that we are having. We would also like to have our
staff cross trained. Do you know somebody good? Or point me in the
right direction?

"Dave Wickert [MSFT]" <dwickert (AT) online (DOT) microsoft.com> wrote