From: Yann Dupont <Yann.Dupont@univ-nantes.fr>
To: Alexandre DERUMIER <aderumier@odiso.com>
Cc: Hannes Reinecke <hare@suse.de>,
Stefan Priebe - Profihost AG <s.priebe@profihost.ag>,
Mark Nelson <mark.nelson@inktank.com>,
ceph-devel@vger.kernel.org, Stefan Majer <stefan.majer@gmail.com>
Subject: Re: Infiniband 40GB
Date: Mon, 04 Jun 2012 11:53:48 +0200 [thread overview]
Message-ID: <4FCC85AC.209@univ-nantes.fr> (raw)
In-Reply-To: <d1ba3477-a5ed-488a-8785-2a37f4301ffc@mailpro>
Le 04/06/2012 11:35, Alexandre DERUMIER a écrit :
> Hi,
> about this:
>>> Turning off Virtualisation extension in BIOS. Don't know why, but it
>>> gaves us crappy performance. We usually put it on, because we use KVM a
>>> lot. In our case, OSD are in bare metal and disabling virtualisation
>>> extension gives us a very big boost.
>>> It may be a BIOS bug in our machines (DELL M610).
>
> It could be related to iommu, if you pass intel_iommu=on in grub.
> I have already had this kind of problem.
>
> When intel_iommu=on, Linux (completely unrelated to KVM) adds a new level
> of protection which didn't exist without an IOMMU - the network card, which
> without an IOMMU could write (via DMA) to any memory location, now is
> not allowed - the card can only write to memory locates which the OS
> wanted it to write. Theoretically, this can protect the OS against
> various kinds of attacks. But what happens now is that every time that
> Linux passes a new buffer to the card, it needs to change the IOMMU
> mappings. This noticably slows down I/O, unfortunately.
>
>
Infortunately, this is not the case. The intel card supports it, but
DELL M160 don't.
And I just checked, ou linux command line don't include intel_iommu=on.
BTW, it seems that turning on virtualization on bios kills performance
on integrated ixgbe driver. Sourceforge one seems less affected. Our
tests were circa kernel 3.2 , it may have changed since.
Cheers,
--
Yann Dupont - Service IRTS, DSI Université de Nantes
Tel : 02.53.48.49.20 - Mail/Jabber : Yann.Dupont@univ-nantes.fr
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2012-06-04 9:53 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-03 8:10 Infiniband 40GB Stefan Priebe
2012-06-03 12:56 ` Mark Nelson
2012-06-04 6:22 ` Hannes Reinecke
2012-06-04 7:26 ` Stefan Priebe - Profihost AG
2012-06-04 7:39 ` Hannes Reinecke
2012-06-04 7:53 ` Stefan Priebe - Profihost AG
2012-06-04 8:02 ` Hannes Reinecke
2012-06-04 8:23 ` Stefan Majer
2012-06-04 9:21 ` Yann Dupont
2012-06-04 9:35 ` Alexandre DERUMIER
2012-06-04 9:53 ` Yann Dupont [this message]
2012-06-04 9:47 ` Amon Ott
2012-06-04 9:58 ` Yann Dupont
2012-06-04 11:40 ` Alexandre DERUMIER
2012-06-04 12:59 ` Mark Nelson
2012-06-04 13:07 ` Alexandre DERUMIER
2012-06-04 13:28 ` Mark Nelson
2012-06-04 15:11 ` Gregory Farnum
2012-06-04 15:34 ` Mark Nelson
2012-06-06 16:05 ` Alexandre DERUMIER
2012-06-06 16:43 ` Mark Nelson
2012-06-04 15:42 ` Stefan Priebe
2012-06-05 7:08 ` Amon Ott
2012-06-05 7:46 ` Stefan Priebe - Profihost AG
2012-06-06 10:48 ` Stefan Priebe - Profihost AG
2012-06-06 10:57 ` Amon Ott
2012-06-06 11:02 ` Stefan Priebe - Profihost AG
2012-06-07 11:33 ` Amon Ott
2012-06-07 12:44 ` Stefan Priebe - Profihost AG
2012-06-05 8:54 ` Stefan Priebe - Profihost AG
2012-06-04 12:28 ` Mark Nelson
2012-06-04 12:34 ` Tomasz Paszkowski
2012-06-04 12:40 ` Mark Nelson
[not found] <a81f3855-1c7d-447b-9bbf-6a891e372909@mailpro>
2012-06-07 3:31 ` Alexandre DERUMIER
2012-06-07 11:25 ` Alexandre DERUMIER
2012-06-07 17:15 ` Mark Nelson
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=4FCC85AC.209@univ-nantes.fr \
--to=yann.dupont@univ-nantes.fr \
--cc=aderumier@odiso.com \
--cc=ceph-devel@vger.kernel.org \
--cc=hare@suse.de \
--cc=mark.nelson@inktank.com \
--cc=s.priebe@profihost.ag \
--cc=stefan.majer@gmail.com \
/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