Viewing 15 posts - 4,051 through 4,065 (of 5,843 total)
1) if you have too few spindles in your IO subsystem, multiple files can actually degrade performance
2) moving data to new filegroups is done via a clustered index, specifying what...
April 9, 2010 at 9:51 am
jswong05 (4/9/2010)
chileu17 (10/21/2008)
April 9, 2010 at 8:12 am
MANU-J. (4/8/2010)
Queuing isn't the best indicator, but Avg Disk Sec/Read and Avg Disk Sec/Write could be used.Usual recommendations for those counters are 10-50ms.
MJ
I posted that 6 hours ago....
April 8, 2010 at 3:02 pm
1) the function could be a big problem here. Not sure of the code for it
2) I doubt the index on your temp table is helpful
3) why use a...
April 8, 2010 at 8:40 am
see my other post on a similar thread as this. ADQL is useless, use avg disk sec/read and /write. Or the sys.dm_io_virtual_file_stats DMV - even better.
April 8, 2010 at 8:36 am
CirquedeSQLeil (4/7/2010)
In 2000 there is also no script...
April 8, 2010 at 8:35 am
ADQL is completely useless as a disk performance metric in the modern age. Use avg disk sec/read and avg disk sec/write instead. Even better, use the SQL Server...
April 8, 2010 at 8:30 am
CirquedeSQLeil (4/7/2010)
TheSQLGuru (4/7/2010)
Paul White NZ (4/7/2010)
TheSQLGuru (4/7/2010)
BAH!! I don't need no stinkin' TESTS!! I KNOW my code will run faster than my client's did!!! :w00t: 😀 😎Superb.
And...
April 7, 2010 at 10:30 pm
Paul White NZ (4/7/2010)
TheSQLGuru (4/7/2010)
BAH!! I don't need no stinkin' TESTS!! I KNOW my code will run faster than my client's did!!! :w00t: 😀 😎Superb.
And before someone...
April 7, 2010 at 7:05 pm
Jeff Moden (4/7/2010)
kramaswamy (4/6/2010)
I think that it could be an interesting summer job for an intern,...
Heh... summer intern evaluating code for performance problems. I believe that probably won't...
April 7, 2010 at 6:22 pm
Marts (4/6/2010)
When I created the new expandable log files, I created two per database that...
April 6, 2010 at 1:25 pm
If the primary is empty then it doesn't matter where it is. 🙂 Just avoid putting things there to avoid fragmentation as it grows back out.
April 6, 2010 at 9:24 am
Good points. In my experience, existing database-hitting code is generally only revisited for 3 reasons:
1) fix a bug
2) add a feature
3) performance becomes unacceptable.
#3...
April 6, 2010 at 8:08 am
That's what I get for trying to rush out a few posts before taking my daughter to school - missed several points! Thanks for clarifications Paul. 🙂
April 6, 2010 at 8:03 am
Marts (4/6/2010)
Hmm chkdsk...
April 6, 2010 at 6:30 am
Viewing 15 posts - 4,051 through 4,065 (of 5,843 total)