Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 9 Feb 2007 14:41:39 -0600
From:      Josh Paetzel <josh@tcbug.org>
To:        freebsd-stable@freebsd.org
Cc:        Artem Kuchin <matrix@itlegion.ru>
Subject:   Re: What is a good choice of sata-ii raid controller for freebsd?
Message-ID:  <200702091441.39848.josh@tcbug.org>
In-Reply-To: <01e601c74c5d$31be19c0$0c00a8c0@Artem>
References:  <00ad01c74b65$79db1710$0c00a8c0@Artem> <45CC72D4.9040104@lxnt.info> <01e601c74c5d$31be19c0$0c00a8c0@Artem>

next in thread | previous in thread | raw e-mail | index | archive | help
On Friday 09 February 2007 09:15, Artem Kuchin wrote:
> Alexander Sabourenkov wrote:
> > Artem Kuchin wrote:
> >> hi!
> >>
> >> I am the original poster of this thread. I have read many
> >> interesting reply during these two days. However, as i said in
> >> the original message due to certification issues i am pretty
> >> limited to INTEL controllers  and i have not seen a single
> >> relevant reply about them. This is interesting. Nobody uses
> >> Intel controllers on FreeBSD or they just suck that much?
> >
> > If you have enough  SATA ports and no need for fancy RAID levels,
> > then my advice is to use gmirror.
> >
> > Hardware RAID1 buys you nothing in perfomance and reliability
> > for a prolonged headache with drivers, bios insanity and
> > monitoring+control tools.
>
> Hm... two points here. I, somehow, do not really believe that
> software raid (gmirror for example) is as reliable as hardware.
> I, deeply inside, believe that i might screw things very badly
> under some heavy load and bad timing conditions. Can't explain it.
> it is religious i guess, but i can be very wrong about this.
>
> However, two perfomance point:
> Under gmirror OS must issue two commands to write to disks and some
> commands to check/set mark that mirrored data is intact.
> Under hardware RAID OS issue sonly one command to write and no
> checking command, since raid controller handles this async.
>
> So, software OS raid must be slower than controller based raid
> anyway.
>
> Am i right here? Any benchmark data on this?
>
> As for reliability of gmirror. I just need to know how it works to
> see for myself that if power turned off in some racing condition
> gmirror will know that disk are out of sync. If it is done than
> gmirror must check sync of disks every read, and that mean two
> command for reading too, which must slow down things. Is it true?
>
> --
> Artem
>

What hardware RAID buys you over gmirror is that you can boot from it. =20
If a drive in the mirror fails the device name available to the OS is=20
still the same.  The FreeBSD loader does not do gmirror, it boots off=20
the raw device, and then gmirror is loaded.  If the drive you are=20
booting off of fails you have to have the BIOS set to boot from the=20
other drive in the mirror, and then you run into 'what is the root=20
device set to in loader.conf' issues.

=46rom a raw speed perspective on an unloaded CPU a 3.0ghz processor is=20
probably just as fast or faster than the embedded processor on a RAID=20
card running at a few hundred mhz.  Sure, once you start talking=20
about CPUs at full load there are advantages to off-loading stuff to=20
a dedicated processor.

=2D-=20
Thanks,

Josh Paetzel



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200702091441.39848.josh>