linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Matt Garman <matthew.garman@gmail.com>
To: Rainer Fuegenstein <rfu@kaneda.iguw.tuwien.ac.at>
Cc: Greg Freemyer <greg.freemyer@gmail.com>, linux-raid@vger.kernel.org
Subject: Re: recommended 4port SATA controller ?
Date: Thu, 17 Sep 2009 16:52:40 -0500	[thread overview]
Message-ID: <20090917215240.GA418@sewage> (raw)
In-Reply-To: <4AB24BFC.6070800@kaneda.iguw.tuwien.ac.at>

On Thu, Sep 17, 2009 at 04:47:24PM +0200, Rainer Fuegenstein wrote:
> the plan is to replace this board with a more power-saving one as
> soon  as the new intel atom plattfrom (pinewood? pinetrail?) comes
> out in  2010/Q1 (hard to tell which  expansion slots the boards
> will have then,  but I bet on PCI).  speed isn't an issue, but low
> power consumption would be nice (but not  easy to find in
> datasheets etc.) since the server is supposed to be  powered by
> photovoltaic panels.

As others have mentioned, when doing SATA over PCI, the bottleneck
will be the PCI bus.  IIRC, SATA2 spec supports up to 300 MB/s, and
PCI is 133 (or 150) MB/s.  Realistically, modern spinning hard
drives usually max out around 100 MB/s, so just two drives can
saturate your PCI bus.  I think those numbers are reasonably close.

But, if, for example, you are serving the data strictly over the
network (i.e. a NAS box), a single Gigabit ethernet connection tops
out at rougly 125 MB/s.  So in this case the network performance
would shadow the reduced performance of the drives on the PCI bus.
(Unless your network controller is also on the PCI bus, then you're
hosed!  It's always good to check the block diagram of the
motherboard.)

Anyway, you asked for four ports, but how about eight?  I'm
surprised no one has mentioned this yet---I'd consider it a bit of a
"classic": Supermicro AOC-SAT2-MV8.

    http://www.newegg.com/Product/Product.aspx?Item=N82E16815121009

Eight SATA ports.  It's PCI-X, but I've used mine in regular 32-bit
PCI slots on several different motherboards without any problems.
Plenty of anecdotal evidence on the net as well of people using this
card in 32-bit PCI slots.  Linux support is good (personal
experience plus anecdotal).

I can't speak to its power consumption; there are no heatsinks on
the board, that should mean something.  I don't know, but I would
guess that the PCI bus wasn't designed to provide a lot of power
anyway.

Hope that helps!
Matt


  reply	other threads:[~2009-09-17 21:52 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-09-17 11:44 recommended 4port SATA controller ? Rainer Fuegenstein
2009-09-17 11:57 ` Majed B.
2009-09-17 12:49 ` Greg Freemyer
2009-09-17 13:04   ` Jon Lewis
2009-09-17 13:17     ` Max Waterman
2009-09-17 13:28   ` Rui Santos
2009-10-13 21:19     ` Bill Davidsen
2009-09-17 13:40   ` Tapani Tarvainen
2009-09-17 14:47   ` Rainer Fuegenstein
2009-09-17 21:52     ` Matt Garman [this message]
2009-09-17 22:41       ` Kristleifur Daðason
2009-09-17 23:48       ` Rainer Fuegenstein
2009-09-21 16:29         ` Matt Garman
2009-09-22 14:05       ` Matthias Urlichs
2009-09-17 13:43 ` Christian Pernegger
2009-09-17 14:57 ` Jon Hardcastle
2009-09-17 21:37 ` John Bridges
2009-09-17 23:02   ` Thomas Fjellstrom
2009-09-17 23:35     ` Kristleifur Daðason
2009-09-17 23:59       ` Thomas Fjellstrom
2009-09-18 10:58         ` mdraid causing mvsas to lockup? (was: Re: recommended 4port SATA controller ?) Thomas Fjellstrom
2009-09-18 23:02           ` Thomas Fjellstrom
2009-09-21 16:16             ` mdraid causing mvsas to lockup? Thomas Fjellstrom
2009-09-27  3:34               ` Thomas Fjellstrom
2009-09-18  0:23     ` recommended 4port SATA controller ? John Bridges
2009-09-18  0:52       ` John Bridges

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20090917215240.GA418@sewage \
    --to=matthew.garman@gmail.com \
    --cc=greg.freemyer@gmail.com \
    --cc=linux-raid@vger.kernel.org \
    --cc=rfu@kaneda.iguw.tuwien.ac.at \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).