Date: Sun, 11 Feb 2001 20:37:56 -0800 From: "Ted Mittelstaedt" <tedm@toybox.placo.com> To: "Mike Meyer" <mwm@mired.org> Cc: <questions@FreeBSD.ORG> Subject: RE: Problems installing 4.x on large disks Message-ID: <001b01c094ad$91e79420$1401a8c0@tedm.placo.com> In-Reply-To: <14982.29199.754559.647316@guru.mired.org>
next in thread | previous in thread | raw e-mail | index | archive | help
OK Mike,
Since your obviously dying for a comparison, here's the
output of bonnie run on one of the spools on our news server.
This is on a stripe set of 3 9GB SCSI disks that has been
built with the command:
newfs -f 512 -i 2048 /dev/rccd1c
on a ccd that was created with an /etc/ccd.conf file of
ccd1 65536 0 /dev/sd4s1e /dev/sd5s1e /dev/sd6s1e
According to the docs for ccd this is supposed to optimize the ccd
for writing lots and lots of itty-bitty files. Bonnie uses a single
large file so I would think the results would be affected somewhat.
I never did bother experimenting with other values of interleaving in
the ccd to see if it would make a difference.
Needless to say I shut down inn while this was run, and since there was
nothing
else sharing the CPU on the system at the time, the CPU figures are high.
Also,
this spool has been in continuous operation for about 5 months and can vary
from a low of 5% to a high of 95% capacity in a single day. When this was
run
the spool was 86% full so Bonnie's test file was probably fragmented
tremendously.
This system is FreeBSD 2.2.8 by the way.
# bonnie -s 256
File './Bonnie.12717', size: 268435456
Writing with putc()...done
Rewriting...done
Writing intelligently...done
Reading with getc()...done
Reading intelligently...done
Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done...
-------Sequential Output-------- ---Sequential
Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per
Char- --Block--- --Seeks---
Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec
%CPU
256 3820 97.0 8054 88.4 4753 59.5 6434 94.3 9524 27.7 195.3
11.0
#
Here is the dmsg fragment covering the disks.
ahc2 <Adaptec 2940 Ultra SCSI host adapter> rev 1 int a irq 10 on pci0:4:0
ahc2: aic7880 Wide Channel, SCSI Id=7, 16/255 SCBs
ahc2 waiting for scsi devices to settle
ahc2: target 0 Tagged Queuing Device
(ahc2:0:0): "SEAGATE ST39102LW 0005" type 0 fixed SCSI 2
sd4(ahc2:0:0): Direct-Access 8683MB (17783240 512 byte sectors)
ahc2: target 1 Tagged Queuing Device
(ahc2:1:0): "SEAGATE ST39102LW 0005" type 0 fixed SCSI 2
sd5(ahc2:1:0): Direct-Access 8683MB (17783240 512 byte sectors)
ahc2: target 2 Tagged Queuing Device
(ahc2:2:0): "SEAGATE ST39102LW 0005" type 0 fixed SCSI 2
sd6(ahc2:2:0): Direct-Access 8683MB (17783240 512 byte sectors)
Ted Mittelstaedt tedm@toybox.placo.com
Author of: The FreeBSD Corporate Networker's Guide
Book website: http://www.freebsd-corp-net-guide.com
> -----Original Message-----
> From: owner-freebsd-questions@FreeBSD.ORG
> [mailto:owner-freebsd-questions@FreeBSD.ORG]On Behalf Of Mike Meyer
> Sent: Sunday, February 11, 2001 3:06 AM
> To: Ted Mittelstaedt
> Cc: questions@FreeBSD.ORG
> Subject: RE: Problems installing 4.x on large disks
>
>
> Ted Mittelstaedt <tedm@toybox.placo.com> types:
> > > Just the opposite in this case. The SCSI drive was a Seagate Barracude
> > > 9LP on a 7890, 7200 PRM with a 7.4ms seek time. The IDE drive was a
> > > Maxtor DiamondMax plus, 7200rpm with a 9ms seek time. The test
> > > software was bonnie.
> > I'm assuming that since you didn't say AHA2940U2 that this was
> an on-board
> > AIC 7890. Those usually have a BIOS configuration that needs to be set
> > to tell it to negotiate at the highest speed the drive would operate it.
> > I'm not familiar with bonnie to know what it is testing.
>
> All the negotiations were enabled. Bonnie is in the ports tree for you
> to look at. Better yet, run your own tests, and report the results
> back. I believed pretty much what you're saying until I ran my own
> tests - prompted by people saying what I'm telling you now.
>
> <mike
> --
> Mike Meyer <mwm@mired.org>
> http://www.mired.org/home/mwm/
> Independent WWW/Perforce/FreeBSD/Unix consultant, email for more
> information.
>
>
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-questions" in the body of the message
>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?001b01c094ad$91e79420$1401a8c0>