| Author |
Slow SQL 7 stored procs when DB is read-only
|
|
| george.semple@gmail.com 2005-10-28, 11:23 am |
| We have several queries that run very slow on read-only db's but very
fast on non-read only db's. Very slow as in 20-30 minutes vs 15-20
seconds so the difference is very noticable.
Any tips on what to start looking for?
I'm very new to T-SQl so take that into consideration. Thanks
| |
| David Gugick 2005-10-28, 1:23 pm |
| george.semple@gmail.com wrote:
> We have several queries that run very slow on read-only db's but very
> fast on non-read only db's. Very slow as in 20-30 minutes vs 15-20
> seconds so the difference is very noticable.
>
> Any tips on what to start looking for?
>
> I'm very new to T-SQl so take that into consideration. Thanks
What do the execution plan differences show, if anything? Have you
updated the statistics on the read only server?
--
David Gugick
Quest Software
www.imceda.com
www.quest.com
| |
| George 2005-10-28, 8:23 pm |
|
"David Gugick" wrote:
> george.semple@gmail.com wrote:
>
> What do the execution plan differences show, if anything? Have you
> updated the statistics on the read only server?
>
> --
> David Gugick
> Quest Software
> www.imceda.com
> www.quest.com
>
>
I won't be able to compare plans until afterhours tonight (the operation
actually causes timeouts while it is running on the read-only db). I will
update the table stats after comparing the two plans at that time as well.
| |
| David Gugick 2005-10-28, 8:23 pm |
| George wrote:
> "David Gugick" wrote:
>
> I won't be able to compare plans until afterhours tonight (the
> operation actually causes timeouts while it is running on the
> read-only db). I will update the table stats after comparing the two
> plans at that time as well.
You can run an estimated plan instead, which will not execute the SQL
statement. That might give you an intial indication until you have a
chance to run them for real.
--
David Gugick
Quest Software
www.imceda.com
www.quest.com
|
|
|
|