All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Atchley, Scott" <atchleyes-1Heg1YXhbW8@public.gmane.org>
To: Reeted <reeted-9AbUPqfR1/2XDw4h08c5KA@public.gmane.org>
Cc: linux-rdma <linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Subject: Re: IPoIB performance
Date: Wed, 5 Sep 2012 15:46:08 -0400	[thread overview]
Message-ID: <1B8328EC-32D4-4D1D-B62C-96EA0B1B4E74@ornl.gov> (raw)
In-Reply-To: <5047A233.2040602-9AbUPqfR1/2XDw4h08c5KA@public.gmane.org>

On Sep 5, 2012, at 3:04 PM, Reeted wrote:

> On 09/05/12 19:59, Atchley, Scott wrote:
>> On Sep 5, 2012, at 1:50 PM, Reeted wrote:
>> 
>>> 
>>> I have read that with newer cards the datagram (unconnected) mode is
>>> faster at IPoIB than connected mode. Do you want to check?
>> I have read that the latency is lower (better) but the bandwidth is lower.
>> 
>> Using datagram mode limits the MTU to 2044 and the throughput to ~3 Gb/s on these machines/cards. Connected mode at the same MTU performs roughly the same. The win in connected mode comes with larger MTUs. With a 9000 MTU, I see ~6 Gb/s. Pushing the MTU to 655120 (the maximum for ipoib), I can get ~13 Gb/s.
>> 
> 
> Have a look at an old thread in this ML by Sebastien Dugue "IPoIB to 
> Ethernet routing performance"
> He had numbers much higher than yours on similar hardware, and was 
> suggested to use datagram to achieve offloading and even higher speeds.
> Keep me informed if you can fix this, I am interested but can't test 
> infiniband myself right now.

He claims 20 Gb/s and Or replies that one should also get near 20 Gb/s using datagram mode. I checked and datagram mode shows support via ethtool for more offloads. In my case, I still see better performance with connected mode.

Thanks,

Scott--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

      parent reply	other threads:[~2012-09-05 19:46 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-08-29 19:35 IPoIB performance Atchley, Scott
     [not found] ` <FFD82983-4A73-4A52-B6BE-C63DA16A507C-1Heg1YXhbW8@public.gmane.org>
2012-09-05 15:51   ` Christoph Lameter
     [not found]     ` <0000013997217ec3-f6e2593f-2ff8-408d-814e-0345582b31ca-000000-p/GC64/jrecnJqMo6gzdpkEOCMrvLtNR@public.gmane.org>
2012-09-05 17:09       ` Atchley, Scott
     [not found]         ` <3F476926-8618-4233-A150-C5D487B55C68-1Heg1YXhbW8@public.gmane.org>
2012-09-05 18:20           ` Christoph Lameter
     [not found]             ` <0000013997a928ab-36ad5a02-3c82-4daf-8e8a-a86c65e92376-000000-p/GC64/jrecnJqMo6gzdpkEOCMrvLtNR@public.gmane.org>
2012-09-05 18:30               ` Atchley, Scott
     [not found]                 ` <78D8717C-0505-408B-8625-A9124AB33C9E-1Heg1YXhbW8@public.gmane.org>
2012-09-05 19:06                   ` Christoph Lameter
     [not found]                     ` <0000013997d35848-57319f33-839b-4480-a075-53b36f67bfe2-000000-p/GC64/jrecnJqMo6gzdpkEOCMrvLtNR@public.gmane.org>
2012-09-05 19:48                       ` Atchley, Scott
     [not found]                         ` <E88DEA9F-416B-4663-A292-5780DFF9B641-1Heg1YXhbW8@public.gmane.org>
2012-09-05 19:53                           ` Christoph Lameter
2012-09-05 20:12                           ` Ezra Kissel
     [not found]                             ` <5047B23C.6080604-GZvvpLG7cYSVc3sceRu5cw@public.gmane.org>
2012-09-05 20:32                               ` Atchley, Scott
2012-09-05 19:13                   ` Christoph Lameter
     [not found]                     ` <0000013997da4014-44dda0e8-01f3-48b8-b0cd-fe41164d590c-000000-p/GC64/jrecnJqMo6gzdpkEOCMrvLtNR@public.gmane.org>
2012-09-05 19:52                       ` Atchley, Scott
     [not found]                         ` <F90894CF-B55C-4AF7-845C-279FDF44351E-1Heg1YXhbW8@public.gmane.org>
2012-09-05 20:26                           ` Christoph Lameter
2012-09-05 17:52       ` Reeted
2012-09-05 17:50   ` Reeted
     [not found]     ` <504790D8.7010307-9AbUPqfR1/2XDw4h08c5KA@public.gmane.org>
2012-09-05 17:59       ` Atchley, Scott
     [not found]         ` <64A9A0CD-00E0-4455-A641-324FA9BB8BC2-1Heg1YXhbW8@public.gmane.org>
2012-09-05 19:04           ` Reeted
     [not found]             ` <5047A233.2040602-9AbUPqfR1/2XDw4h08c5KA@public.gmane.org>
2012-09-05 19:46               ` Atchley, Scott [this message]

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=1B8328EC-32D4-4D1D-B62C-96EA0B1B4E74@ornl.gov \
    --to=atchleyes-1heg1yxhbw8@public.gmane.org \
    --cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=reeted-9AbUPqfR1/2XDw4h08c5KA@public.gmane.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.