Viewing 15 posts - 991 through 1,005 (of 2,038 total)
UMG Developer (5/11/2009)
... (Edit: After 2 and half hours I killed it, it was on the "Move last fragments up" step.)
Well, errmm... You need bigger hardware! 😀
Seems to depend on...
May 12, 2009 at 2:21 am
UMG Developer (5/11/2009)
Florian Reischl (5/11/2009)
Have a look at the attached statistic results of my tests.Flo,
Which cursor based split solution were you using in those statistics?
Hi UMG
Here is the cursor I...
May 12, 2009 at 2:17 am
Hi Richard
Go to "SQL Server Configuration Manager".
* Select "SQL Server Network Configuration" -> "Protocols for MSSQLSERVER" and enable "Named Pipes" and "TCP/IP" (to be able to connect from another...
May 11, 2009 at 4:59 pm
I have to say the cursors not only win the MAX data-type contest. The set-based solutions have a generally problem with large items within the source text. Have a look...
May 11, 2009 at 4:12 pm
Just two simple thing:
Change the initial database/logfile sizes and change the stupid default database configuration to grow files by 1 MB!!
I will never get tired of tellin this...
Greets
Flo
May 11, 2009 at 4:06 pm
Bob Hovious (5/11/2009)
8.Indentation?! - I will show you how to indent when I indent your skull!
Don't forget, no indentation saves the environment. More spaces need more disk space and more...
May 11, 2009 at 3:59 pm
Hi Paul
I know this sound's strange but all my tests showed, that the set-based solutions using a Tally table run into problems with huge text and large result items.
Remember Phil's...
May 11, 2009 at 3:48 pm
May 11, 2009 at 3:22 pm
Hi Ray
I'm not sure if I understood you. But if you want to update the fields of your @Primary only if they are currently NULL try this:
[font="Courier New"]DECLARE @t TABLE...
May 11, 2009 at 12:50 pm
UMG Developer (5/10/2009)
Yes the data is all text. I'll have to take a look at your proposed solution, but it would need some work to handle all the possible crap...
May 11, 2009 at 12:18 pm
Hi Karl
Thanks for your feedback. No his name is not Nigel 😉
I will inform him that there shouldn't be any problems with online index rebuilds.
Greets
Flo
May 11, 2009 at 9:45 am
Hi Jeff
I saw your question and you are correct. If the data are imported by a file this would be the best approach.
But I already started with my solution and...
May 10, 2009 at 4:11 pm
Hi UMG
For now just a simple approach.
Since your source data are not enclosed with a leading/trailing delimiter I took my UNION ALL Tally version to do the initial split. This...
May 10, 2009 at 3:54 pm
Hi UMG
Maybe I have an idea. Just for ensure. Your data are text data? No binary.
Greets
Flo
May 10, 2009 at 3:11 pm
Viewing 15 posts - 991 through 1,005 (of 2,038 total)