Viewing 15 posts - 1,066 through 1,080 (of 1,162 total)
Yes, not sure what you're trying to accomplish by applying a power to any negative number as even with whole integers the result flip-flops between a positive or negative result...
March 5, 2010 at 8:35 am
Yikes. Having users in the business that have permission to manually update data outside of the confines of the application is a massive problem.
The only real way to maintain integrity...
March 4, 2010 at 10:11 am
The only possible reason I can think for this is that you actually have values that are 'NULL' as character value rather than being NULL in the database sense. Do...
March 4, 2010 at 6:18 am
The article you quoted seems to suggest that CU5 includes a 'real' fix:
http://support.microsoft.com/kb/976603/
Microsoft have now released this fix which is a thorough solution to this problem, which resolves the original...
March 3, 2010 at 8:54 am
Thanks Gail, good to know. FYI, I tried this scenario in my local 2008 SP1 instance and it doesn't appear to change the plan that's produced compared to 2k5
March 3, 2010 at 7:30 am
I don't disagree that it's difficult given SQL's architecture, but in this case (and it's not an unusual case) you're better off if plans aren't reused at all.
It was a...
March 3, 2010 at 6:28 am
Are you saying that the execution plan shows a Clustered Index Scan rather than a Clustered Index Seek?
This is a common problem with catch all queries.
Even though this parameter is...
March 3, 2010 at 4:15 am
Perhaps, but you do pay a price in poorer performance and reduced features and flexibility.
Could you point me in the direction of a source for MERGE having better performance than...
March 1, 2010 at 9:40 am
From BOL:
http://technet.microsoft.com/en-us/library/bb510625.aspx
Because the MERGE statement performs a full table scan of both the source and target tables, I/O performance can be affected when using the TOP clause to modify a...
March 1, 2010 at 9:10 am
I honestly cannot see them even deprecating UPDATE FROM, let alone removing it from the product. I think it's crazy to avoid using it in code just because someone said...
March 1, 2010 at 8:37 am
A tally table approach may be faster.
E.g. create a tally table as in this article:
http://www.sqlservercentral.com/articles/T-SQL/62867/
Create a lookup table for the find and replace strings:
CREATE TABLE Replace_Lookup...
February 26, 2010 at 2:54 am
Did you mean to join to the inserted table in your queries or do you actually have a table called "inseted"?
Also, you need to be careful that your set operations...
February 24, 2010 at 6:53 am
Pretty sure a backup/restore from SQL 2000 to SQL 2008 cannot convert an ntext datatype to varchar(2000).
text, ntext and image are all deprecated in SQL 2008, but they still exist....
February 22, 2010 at 10:19 am
I thought using a column in the order by that's included in the partition is irrelevent as the partitioned column by definition will be the same value for all rows...
February 22, 2010 at 9:15 am
The answer is no, not without breaking it into smaller batches.
You submit a query and then get no response back from SQL until it's performed the operation. It's possible you...
February 17, 2010 at 10:31 am
Viewing 15 posts - 1,066 through 1,080 (of 1,162 total)