OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
 
Subject: Re: SCSI, U2W in particular?, and poor performance...
From: Justin Robertson (zululinux.com)
Date: Sun Mar 05 2000 - 14:20:32 CST

FINALLY! A real response. This is what I was origionally thinking, glad to
see you agree.

Just for everbody who still disagrees; here's some interesting stuff.

IDE Machine
(sync)

# grep 'hda' /etc/mtab
/dev/hda1 / ext2 rw,sync 0 0

# /usr/bin/time ./smtp-source -l 15024 -m 1000 -f rootlocalhost -t
rootlocalhost -d 127.0.0.1:25
0.27user 0.39system 0:50.27elapsed 1%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (137major+18minor)pagefaults 0swaps

SCSI Machine
(async on - making a point)

# grep 'sda' /etc/mtab
/dev/sdb1 / ext2 rw,errors=remount-ro,errors=remount-ro 0 0

# /usr/bin/time ./smtp-source -l 15024 -m1000 -f rootlocalhost -t
rootlocalhost -d 127.0.0.1:25
0.66user 1.82system 4:38.42elapsed 0%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (142major+22minor)pagefaults 0swaps

I'd be willing to bet that Dylan is right on this one.

Justin Robertson
<zululinux.com>

On Sun, 5 Mar 2000, Dylan Griffiths wrote:

> Sander Steffann wrote:
> > Why is everybody doing so difficult about this instead of trying to help?
> > The guy mentioned that he replaced the SCSI disk in the fast machine with an
> > IDE disk, and is still noticing a difference in favor of the IDE disk. This
> > means that on the same installation/configuration, with the only difference
> > the SCSI/IDE disk, there is a difference. He only asked if anybody had an
> > idea what could cause this.
>
> It could be a problem with the Linux SCSI subsystem. While stable, is it
> somewhat evil code as of the 2.2.x kernel revisions. Much works has been
> done in the 2.3.x development series to clean up the SCSI subsystem (with
> good results). Also, the SCSI HDs likely support TCQ (tagged command
> queing), but this is not enabled by default in the Linux kernel for
> compatibility. I suggest he tunes the Linux kernel SCSI AIC7xxx driver
> parameters before he does anything else, as the "most compatible" SCSI mode
> is slower than the "best speed" mode of IDE (which is also default).
>
> Otherwise, this is not a postfix issue, nor does it nessitate further
> discussion/flaming on this list.
>
> --
> Hi! I'm a .signature virus! Copy me into your ~/.signature to help me
> spread!
>