Viewing 15 posts - 21,751 through 21,765 (of 22,184 total)
The short answer is "as often as possible." What this translates to varies widely based on system, space, etc.. The reason is pretty much as you state. You need to...
June 8, 2007 at 4:36 am
It's funny you say that. The reason I posted an XML plan was that the graphical plans were acting all wonky with that script. There's a mystery I'm going to...
June 8, 2007 at 4:29 am
Basically, yes. I'm working on something and rather than simply caution that differences could arise, I'd like to show one.
June 7, 2007 at 10:21 am
Crud. Now parameter sniffing is taking part. If I take away the constant values & substitute a parameter I get clustered scans for everything...
**^& %^*%&% *%&$^& $*^%& ##%#!
And I...
June 7, 2007 at 8:31 am
I'm a little spacy this morning (production problems overnight), so maybe I'm doing something stupid here.
In desperation to try to find an answer this question, I hit up a...
June 7, 2007 at 8:13 am
Two identical plans:
<
ShowPlanXML xmlns="http://schemas.microsoft.com/sqlserver/2004/07/showplan" Version="1.0"
June 7, 2007 at 7:07 am
Ha!
Well, I'm not going for messed up, but it's worth a shot. I'll let you know.
June 6, 2007 at 9:59 am
Everything the Guru said and, check the use of transactions in your process. Something seems to be doing some sort of lock escalation within the transaction because blocking, by itself, doesn't...
June 6, 2007 at 8:53 am
OK. Weak theory completely destroyed by reality. You can't drop statistics created by indexes and the other statistics just don't seem to be affecting the plan much. More ideas needed.
June 6, 2007 at 6:41 am
Something like this?
UPDATE Receipts
SET DateEnd = u.ExpiryDate
FROM UserRoles u
INNER JOIN Receipts r
ON u.UserId = r.UserId
WHERE u.RoleId = 5
June 6, 2007 at 6:39 am
I'm thinking I'll try dropping the statistics on a table and see what happens. I'm trying to get a different plan, not just subtle differences in the plan (different row counts...
June 6, 2007 at 6:18 am
Hey,
I saw the test you'd outlined so I tried it on my laptop:
WHILE LOOP TEST...
1000000 The final count (just to show work was done)
970 Duration in milliseconds
------------------------------------------------------------------------------
SET BASED...
June 6, 2007 at 5:19 am
I worked for one of the software vendors for healthcare products. They're out of business now. We actually had very good developers, but a CEO that thought that IT knowledge...
June 1, 2007 at 7:47 am
It comes down to the classic DBA answer, it depends. I've primarily done work for internal applications, code that runs in-house for a company doing X. Platforms I've done that...
June 1, 2007 at 5:32 am
Viewing 15 posts - 21,751 through 21,765 (of 22,184 total)