Viewing 15 posts - 16 through 30 (of 108 total)
truncate_only does not consolidate the data, so only space at the end of the file is released.
May 18, 2006 at 10:07 am
I know why software quality is so low.
It is due to the charade that software is licensed, not sold. This legal fiction is the reason utter crap is allowed into...
May 8, 2006 at 8:22 am
First, determine where the time is spent - on the db server or not (network, client, etc.). Was your QA trial run from the same box as the Web app?
If...
May 4, 2006 at 1:45 pm
With all the tables being added, log shipping sounds more like your best solution. It need not be the built-in variety either - you can roll your own pretty easily.
May 4, 2006 at 1:04 pm
The round-robin/proportional fill doesn't work right for tempdb unless all the files are the same size. If the file that the server chooses to place an object on is not...
May 2, 2006 at 3:05 pm
I ran into that error with a db2 linked server; instead of server.database.schema.table, what worked was server..schema.table, leaving out the database.
April 21, 2006 at 8:08 pm
Converting from datetime to varchar and back would be an awful thing to do for looking at a million rows, but performed once per call to modify a stored procedure...
April 13, 2006 at 4:20 am
I will bet 25c on:
EXEC ('DBCC SHOWCONTIG ([''FAC treaties$'']) WITH ALL_INDEXES, TABLERESULTS, NO_INFOMSGS')
(those are two single quotes, not double quotes)
April 13, 2006 at 4:11 am
Try using UPDATE STATISTICS tablename WITH FULLSCAN. sp_updatestats uses either the last sample size, or the default (calculated internally). Then dbcc freeprocache to toss out existing plan.
April 13, 2006 at 4:04 am
The truncate_only option does not attempt to relocate data, so only releases space that happens to be at the end of the file. Also, shrinkdatabase will not reduce the size...
April 13, 2006 at 3:25 am
For financial transactions, even at the low/medium rate of 1 per 3 seconds, I still would not settle for anything less than 2 processors, 2 GB (4 better), and at...
April 13, 2006 at 3:06 am
Yes, separate threads are used for separate files. Yes, spindles = performance. The situation you are in is difficult to guess at; trial and measurement is the only reliable method....
April 11, 2006 at 8:20 am
I am curious if you did it - which did you do first? Did it seem to make any significant difference in performance?
April 7, 2006 at 8:51 pm
If you really want to analyze the impact on the server - your db vs. others, get ahold of the PSSDiag and Read80Trace utilities from Microsoft (search the KB). They...
April 7, 2006 at 8:37 pm
I think all of your own suggestions are good ideas. I like them roughly in the reverse order given...5-4-3-2-1. Sell number 5 to management by casting it as a high...
April 7, 2006 at 8:23 pm
Viewing 15 posts - 16 through 30 (of 108 total)