Viewing 15 posts - 1 through 15 (of 26 total)
It's good to see such sensible advice here. Yes, a table should always have a clustered index, and IMO an Identity helps solve many of the issues with page splits...
May 24, 2010 at 2:06 am
I absolutely agree with the previous post about set single user. If you have to do this on a per database level then it is totally low rent.
I had...
October 13, 2006 at 2:35 am
Or better yet
update SequenceTable
set @NewSequenceNumber = CurrentSequenceNumber = CurrentSequenceNumber + 1
where SequenceYear = year(getdate())
Having said that, central locations for keys are a concurrency nightmare for anything that needs to...
May 17, 2006 at 8:55 am
As for why cursors are percieved as bad in SQL Server, it's simply about performance. SQL is optimised for set based operations. 'Never use cursors' is not a rule, but...
January 17, 2006 at 7:12 am
I think it's worth making the point that this approach does nothing to combat what is fundamentally bad about cursors. All we have done here is created a cursor, without using...
January 17, 2006 at 4:36 am
I for one would be very interested in attending a session by Fabian. It's always good to test out some challenging ideas with others in the same field, controversial...
January 10, 2006 at 4:30 am
Sorry, I had a typo. Here is an example:
CREATE
TABLE ids ( id int)
GO
January 9, 2006 at 4:09 am
Here is a better approach.
CREATE PROCEDURE Get_New_ID
@Prefix varchar(2)
AS
DECLARE @New_id INT --Assuming this is an int
UPDATE TABLE_ID
SET @new_id = currId = @currentId + 1
January 9, 2006 at 4:06 am
One thing to note.
@@Datefirst is affected by things like the defaultlanguage property of your language.
I believe that for the US @@Datefirst defaults to 7, whereas if you set up...
November 3, 2005 at 5:38 am
Another reason maybe that you have a bottleneck elsewhere that is slowing things down.
Have had a look at the following counters:
Logical Disk: Avg. Disk sec/Transfer -
Logical...
November 2, 2005 at 5:42 am
SQL Server 2000 should run just fine on Win2k3.
Do you have any more information regarding the configuration of the server. Is there nothing all in the errorlog?
November 1, 2005 at 4:23 am
Paul's approach can be used highly effectively if you have a table full of integers. Just use Top 1 and the filter criteria, and you can use it whereever you...
November 1, 2005 at 4:17 am
Hold on a second.
You have configured cost threshold for parallelism to be 1. You also have prioirty boost tuned on
This means that any query that will take more than a...
October 31, 2005 at 2:42 am
I would start turning off priority boost (a strange thing to turn on), and then set max and min server memory back to the defaults (just to get things on...
October 28, 2005 at 4:47 pm
Hold on.
These tables have 1000 rows? Really? How large is the database (run sp_spaceused when in the database and send the output.
I very much doubt that sql will be...
October 28, 2005 at 8:17 am
Viewing 15 posts - 1 through 15 (of 26 total)