Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1265320

Re: linux-next network throughput performance regression

Show all headers | View raw

On 9 November 2015 at 13:23, David Miller <davem@davemloft.net> wrote:
> From: Dexuan Cui <decui@microsoft.com>
> Date: Mon, 9 Nov 2015 03:11:35 +0000
>
>>> -----Original Message-----
>>> From: David Miller [mailto:davem@davemloft.net]
>>> Sent: Monday, November 9, 2015 10:53
>>> To: Dexuan Cui <decui@microsoft.com>
>>> Cc: eric.dumazet@gmail.com; dsa@cumulusnetworks.com; Simon Xiao
>>> <sixiao@microsoft.com>; netdev@vger.kernel.org; Haiyang Zhang
>>> <haiyangz@microsoft.com>; linux-kernel@vger.kernel.org;
>>> devel@linuxdriverproject.org
>>> Subject: Re: linux-next network throughput performance regression
>>>
>>> From: Dexuan Cui <decui@microsoft.com>
>>> Date: Mon, 9 Nov 2015 02:39:24 +0000
>>>
>>> >> Throughput on a single TCP flow for a 40G NIC can be tricky to tune.
>>> > Why is a single TCP flow trickier than multiple TCP flows?
>>> > IMO it should be easier to analyze the issue of a single TCP flow?
>>>
>>> Because a single TCP flow can only use one of the many TX queues
>>> that such modern NICs have.
>>>
>>> The single TX queue becomes the bottleneck.
>>>
>>> Whereas if you have several TCP flows, all of them can use independant
>>> TX queues on the NIC in parallel to fill the link with traffic.
>>>
>>> That's why.
>>
>> Thanks, David!
>> I understand 1 TX queue is the bottleneck (however in Simon's
>> test, TX=1 => 36.7Gb/s, TX=8 => 37.7 Gb/s, so it looks the TX=1 bottleneck
>> is not so obvious).
>> I'm just wondering how the bottleneck became much narrower with
>> recent linux-next in Simon's result (36.7 Gb/s vs. 18.2 Gb/s). IMO there
>> must be some latency somewhere.
>
> I think the whole thing here is that you misinterpreted what Eric said.
>
> He is not arguing that some regression did, or did not, happen.
>
> He instead was making the basic statement about the fact that due to
> the lack of paralellness a single stream TCP case is harder to
> optimize for high speed NICs.
>
> That is all.

We recently had a regression tracked down in a similiar area that was
because of link order.

Dave.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

linux-next network throughput performance regression Simon Xiao <sixiao@microsoft.com> - 2015-11-06 21:40 +0100
  Re: linux-next network throughput performance regression David Ahern <dsa@cumulusnetworks.com> - 2015-11-06 21:50 +0100
    Re: linux-next network throughput performance regression David Ahern <dsa@cumulusnetworks.com> - 2015-11-06 22:40 +0100
      Re: linux-next network throughput performance regression Eric Dumazet <eric.dumazet@gmail.com> - 2015-11-07 20:40 +0100
        Re: linux-next network throughput performance regression Eric Dumazet <eric.dumazet@gmail.com> - 2015-11-07 21:00 +0100
          RE: linux-next network throughput performance regression Simon Xiao <sixiao@microsoft.com> - 2015-11-09 21:30 +0100
            Re: linux-next network throughput performance regression Eric Dumazet <eric.dumazet@gmail.com> - 2015-11-10 00:10 +0100
        Re: linux-next network throughput performance regression David Miller <davem@davemloft.net> - 2015-11-09 04:00 +0100
          Re: linux-next network throughput performance regression David Miller <davem@davemloft.net> - 2015-11-09 04:30 +0100
            Re: linux-next network throughput performance regression Dave Airlie <airlied@gmail.com> - 2015-11-09 04:40 +0100
            Re: linux-next network throughput performance regression Tom Herbert <tom@herbertland.com> - 2015-11-09 06:40 +0100

csiph-web