Viewing 15 posts - 166 through 180 (of 335 total)
Bill,
The reason you don't want to shrink your log file is because it is expensive to grow again. If your log file is 2.5 gig and you shrink it to...
March 29, 2006 at 3:20 pm
Have you looked at the index fragmentation on the tables. What about statistics. Additional indexes to speed up deletes. I assume that there is a clustered index if not I...
March 29, 2006 at 3:06 pm
I would run my own reindex script on the tables and see if space changes. Sp_spaceused will give you the size of the object and the index size. Also, I...
March 16, 2006 at 3:05 pm
I haven't looked in the scripts for a while, but I think what your looking for is a kill scripts. Here is one that takes a parameter of database name...
March 7, 2006 at 3:15 pm
Without seeing the execution plans my guess right now would be parallelism. To test on production, I would set cost threshold for parallelism to 12 - 15. That is normally...
February 23, 2006 at 8:22 am
How many processors are on the 2 servers. prod and test. If you run sp_configure what is the cost threshold for parallelism? min memory per query (KB)? Are the indexes the...
February 22, 2006 at 12:54 pm
Are the objects being referenced fully qualified (owner.objectname). Is the same login executing the command from both the server and client. What net library is each using to connect. Have...
February 22, 2006 at 12:48 pm
Try setting the user to someone else other than yourself or sa, and see if that takes. If it takes, then set it to sa.
Tom
February 22, 2006 at 12:40 pm
Why not put a trigger on the table to capture when the change is happening and who is executing it. You could also use the trigger to rollback any transaction...
February 17, 2006 at 7:22 am
At least for testing purposes you can put a hint in the query to force it to use the index that you want and see what the performance is. Does...
February 13, 2006 at 2:12 pm
What is the code? It sounds like maybe a different execution plan is being chosen on each run. Is this dynamic sql? Hardware specs don't tell us that your selection...
January 30, 2006 at 12:59 pm
I would also look at the execution plan on the stored procedure to see if a different index could be utilized. In general I size tempdb at 2gig and allow...
January 10, 2006 at 12:27 pm
This might work for you.
SELECT
rm.rmsacctnum AS [Rms Acct Num]
, rf.rmstranamt10 as total_10
, rf.rmstranamt
, (rf.rmstranamt - rf.rmstranamt10) AS [Current Balance]
FROM RMASTER rm
INNER JOIN (
SELECT RMSFILENUM,
RMSTRANCDE,
SUM(rmstranamt)...
January 10, 2006 at 10:04 am
We use a calendar table. It lists every day in the year and its corresponding values.
day fiscal week calendar week iso date fiscal month calendar month
ie. 1-1-2006 40 1 ...
January 5, 2006 at 10:16 am
Contrary to what was stated earlier there is extra overhead to checking the existence before the delete. Your executing a statement twice. Potentially flushing things out of cache. The bigger question is why. ...
January 3, 2006 at 10:08 am
Viewing 15 posts - 166 through 180 (of 335 total)