Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 22 Jul 2006 22:51:59 -0500
From:      "Nikolas Britton" <nikolas.britton@gmail.com>
To:        "Chad Leigh -- Shire.Net LLC" <chad@shire.net>
Cc:        "freebsd-questions@FreeBSD.org" <freebsd-questions@freebsd.org>
Subject:   Re: Areca RAID Card.
Message-ID:  <ef10de9a0607222051p32dad354ua4ffad9f86b9f3@mail.gmail.com>
In-Reply-To: <ef10de9a0607221813v5bfc960crae6fc15c584eb06b@mail.gmail.com>
References:  <ef10de9a0607220314j6f55c2d5x6f3219e6227017e9@mail.gmail.com> <B353A74B-DA62-4633-B5C5-16FD2CA503A5@shire.net> <ef10de9a0607221813v5bfc960crae6fc15c584eb06b@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On 7/22/06, Nikolas Britton <nikolas.britton@gmail.com> wrote:
> On 7/22/06, Chad Leigh -- Shire.Net LLC <chad@shire.net> wrote:
> >
> > On Jul 22, 2006, at 4:14 AM, Nikolas Britton wrote:
> >
> > > The motherboard I just bought, SuperMicro X7DBE, has both PCI-X
> > > 133MHz/64-bit and PCI-Express 8x slots, I can't decide what version of
> > > the Areca card to get... I'd like to see some benchmarks of the
> > > ARC-11xx (PCI-X) Vs. the ARC-12xx (PCI-Express).
> > >
> > > The PCIe device has a faster bus (PCIe 8x = 2000MB/s) but PCI-X is
> > > tried and true and not too shabby (PCI-X 133/64 = 850MB/s) ether.
> > >
> > > I have the option of ether a 1130ML (Infiniband connections) or a plan
> > > jane 1230. I've had troubles with SATA cables in the passed so the
> > > 1130ML is very desirable from this stand point. Another thing I'm
> > > worried about is the 1230 will have to much weight on the PCIe 8x slot
> > > because of all the SATA cables. and routing them all is a pain. Does
> > > anyone have a source for an ARC-1230ML? On the other hand I've never
> > > tried the latching SATA cables yet... but the ARC-1130 is $40
> > > cheaper...
> >
> > I was just looking at the difference between the 1130 and the
> > 1130ML.  ML cables are EXPENSIVE and look heavier than 4 normal
> > cables...
>
> But they latch on and you only need 1 ML cable for every 4 SATA
> cables. The PCI-X card/slot should be sturdy enough but I don't think
> PCIe is, I've played with PCIe 1x cards and their super small...
> picture a normal low profile PCI card, now take half that. Maybe they
> don't make them (ARC-12xxML) for this reason.
>
> > I just ordered an 1120 from <http://www.topmicrousa.com/controllers--
> > tekram.html> and they were the cheapest I've seen.
> >
> > Will be needing an 1130 myself soon I think.
> >
> > Can't help you with the 1130 vs 1230.  I would think the PCIe would
> > be the way to go for future proofing your investment.
> >
>
> Yes I think your right here. If you look at the "ATTO STRs and cache
> transfer rates"[1] the ARC-1120 (PCI-X) is up against the bus limit.
> theoretically the 1120/1220 could do up to 2400MB/s (8 drives *
> SATA-II transfer limit of 300MB/s).
>
> "The results of the Areca ARC-1120 in the RAID 0 tests cleary show
> this adapter does not have any trouble with ATTO's tiny dataset.
> Floating high above the crowd, the ARC-1120 has a perfect view on the
> struggles of the other adapters. Exceeding 750MB/s, the transfer rates
> from the Areca ARC-1120 are almost equal to the effective bandwidth of
> the 133MHz PCI-X bus." [1]
>
> [1] http://tweakers.net/reviews/557/18
>

The only difference between the ARC-11xx and the ARC-12xx's Intel
XScale processor is the IOP333 on the ARC-12xx* has a "PCI Express to
PCI-X Bridge"... Both chips are otherwise identical and both chips use
PCI-X133 internally. Also it appears the IOP333's internal bus
operates at 333MHz while the IOP331 operates at 266MHz but I can't
confirm this, it's possible older IOP331's worked at 266MHz but new
ones are 333MHz now.

ftp://download.intel.com/design/iio/prodbref/25341301.pdf
ftp://download.intel.com/design/iio/prodbref/30658301.pdf

*The ARC-1210 uses a IOP332.


-- 
BSD Podcasts @:
http://bsdtalk.blogspot.com/
http://freebsdforall.blogspot.com/



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ef10de9a0607222051p32dad354ua4ffad9f86b9f3>