I would guess that the processor speed had more to do with it. Yes, the
older box is dual, but if you're only testing a single page at a time
your're not getting as much advantage out of it. You could try removing
half the RAM i the new box and try again...
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, October 03, 2000 3:52 PM
To: CF-Talk
Subject: Benchmarking Servers
Maybe I shouldn't be surprised, but I am.
I'm developing a small server farm. Just
today finally got server #2 up for testing.
Old Server: Dual Pentium II 450, NT4sp6a,
Mirrored Seagate SCSI HD. CF4.01, 256MB Ram
(only 212 MB used at the time)
New Server: Single Pentium III 733, NT4sp6a,
Single IBM ATA/66 HD. CF 4.51, 512MB Ram
(So-called cheap server, $1500)
Both servers access SQL7 on a third server
on the LAN.
For the identical tasks pulling data out of a
cached query and displaying it on a page.
New server took somewhat less than one-half
the time as Old Server (170ms vs 80ms).
Of course, in this case the hard drives
entered into the equation only minimally.
Guess this only reinforces the prevailing
opinion in some circles to go with more RAM
instead of fast hard drives (if this can
do the job in your specific case).
best, paul
==============================
Paul Smith, Web/Database Droid
A: SupportNet, Inc, 3871 Piedmont Ave, Oakland, CA 94611
(There, there, there; there's, there, there....)
P: (510) 763-2358
C: (510) 205-6755
F: (510) 763-2370
E: [EMAIL PROTECTED]
W: http://www.support.net
----------------------------------------------------------------------------
--
Archives: http://www.mail-archive.com/[email protected]/
To Unsubscribe visit
http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or
send a message to [EMAIL PROTECTED] with 'unsubscribe' in
the body.
------------------------------------------------------------------------------
Archives: http://www.mail-archive.com/[email protected]/
To Unsubscribe visit
http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or send a
message to [EMAIL PROTECTED] with 'unsubscribe' in the body.
