General list for discussing Bufferbloat
 help / color / mirror / Atom feed
From: Neal Cardwell <ncardwell@google.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Jonathan Morton <chromatix99@gmail.com>,
	Michael Welzl <michawe@ifi.uio.no>, David Ros <dros@simula.no>,
	bloat <bloat@lists.bufferbloat.net>,
	 Yuchung Cheng <ycheng@google.com>, Wei Wang <weiwan@google.com>
Subject: Re: [Bloat] benefits of ack filtering
Date: Thu, 30 Nov 2017 09:51:56 -0500	[thread overview]
Message-ID: <CADVnQyn0x3s6vCg7RtQ=2tWHTNRJDQsm4qu+6PnPAPfv6nyxGw@mail.gmail.com> (raw)
In-Reply-To: <1512037480.19682.10.camel@gmail.com>

[-- Attachment #1: Type: text/plain, Size: 3786 bytes --]

On Thu, Nov 30, 2017 at 5:24 AM, Eric Dumazet <eric.dumazet@gmail.com>
wrote:

> I agree that TCP itself should generate ACK smarter, on receivers that
> are lacking GRO. (TCP sends at most one ACK per GRO packets, that is
> why we did not feel an urgent need for better ACK generation)
>
> It is actually difficult task, because it might need an additional
> timer, and we were reluctant adding extra complexity for that.
>

How about just using the existing delayed ACK timer, and just making the
delayed ACK logic a bit smarter? We could try using the existing logic and
timers, but using something adaptive instead of the magic "2" MSS received
to force an ACK.


> An additional point where huge gains are possible is to add TSO
> autodefer while in recovery. Lacking TSO auto defer explains why TCP
> flows enter a degenerated behavior, re-sending 1-MSS packets in
> response to SACK flood.
>

Yes, agreed. I suspect there is some simple heuristic that could be
implemented to allow TSO deferral for most packets sent in recovery. For
example, allowing TSO deferral once the number of packet bursts (TSO skbs)
sent in recovery is greater than some threshold. Perhaps TSO deferral would
be fine in Recovery if we have sent, say, 10 skbs, because at that point if
the ACK stream from the original flight dries up due to massive/tail loss,
we have probably sent enough data in the new flight in Recovery to ensure
some kind of ACKs come back to keep the ACK clock going.

neal


>
>
> On Thu, 2017-11-30 at 09:48 +0200, Jonathan Morton wrote:
> > I do see your arguments.  Let it be known that I didn't initiate the
> > ack-filter in Cake, though it does seem to work quite well.
> > With respect to BBR, I don't think it depends strongly on the return
> > rate of acks in themselves, but rather on the rate of sequence number
> > advance that they indicate.  For this purpose, having the receiver
> > emit sparser but still regularly spaced acks would be better than
> > having some middlebox delete some less-predictable subset of them.
> > So I think BBR could be a good testbed for AckCC implementation,
> > especially as it is inherently paced and thus doesn't suffer from
> > burstiness as a conventional ack-clocked TCP might.
> > The real trouble with AckCC is that it requires implementation on the
> > client as well as the server.  That's most likely why Google hasn't
> > tried it yet; there are no receivers in the wild that would give them
> > valid data on its effectiveness.  Adding support in Linux would help
> > here, but aside from Android devices, Linux is only a relatively
> > small proportion of Google's client traffic - and Android devices are
> > slow to pick up new kernel features if they can't immediately turn it
> > into a consumer-friendly bullet point.
> > Meanwhile we have highly asymmetric last-mile links (10:1 is typical,
> > 50:1 is occasionally seen), where a large fraction of upload
> > bandwidth is occupied by acks in order to fully utilise the download
> > bandwidth in TCP.  Any concurrent upload flows have to compete with
> > that dense ack flow, which in various schemes is unfair to either the
> > upload or the download throughput.
> > That is a problem as soon as you have multiple users on the same
> > link, eg. a family household at the weekend.  Thinning out those acks
> > in response to uplink congestion is a solution.  Maybe not the best
> > possible solution, but a deployable one that works.
> > - Jonathan Morton
> > _______________________________________________
> > Bloat mailing list
> > Bloat@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/bloat
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>

[-- Attachment #2: Type: text/html, Size: 5100 bytes --]

  parent reply	other threads:[~2017-11-30 14:52 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-28 21:48 Dave Taht
2017-11-29  6:09 ` Mikael Abrahamsson
2017-11-29  9:34   ` Sebastian Moeller
2017-11-29 12:49     ` Mikael Abrahamsson
2017-11-29 13:13       ` Luca Muscariello
2017-11-29 14:31         ` Mikael Abrahamsson
2017-11-29 14:36           ` Jonathan Morton
2017-11-29 15:24             ` Andrés Arcia-Moret
2017-11-29 15:53           ` Luca Muscariello
     [not found]             ` <CAJq5cE3qsmy8EFYZmQsLL_frm8Tty9Gkm92MQPZ649+kpM1oMw@mail.gmail.com>
2017-11-29 16:13               ` Jonathan Morton
2017-11-30  7:03             ` Michael Welzl
2017-11-30  7:24               ` Dave Taht
2017-11-30  7:45               ` Dave Taht
2017-11-30  7:48               ` Jonathan Morton
2017-11-30  8:00                 ` Luca Muscariello
2017-11-30 10:24                 ` Eric Dumazet
2017-11-30 13:04                   ` Mikael Abrahamsson
2017-11-30 15:51                     ` Eric Dumazet
2017-12-01  0:28                     ` David Lang
2017-12-01  7:09                       ` Jan Ceuleers
2017-12-01 12:53                         ` Toke Høiland-Jørgensen
2017-12-01 13:13                           ` [Bloat] [Make-wifi-fast] " Кирилл Луконин
2017-12-01 13:22                             ` Luca Muscariello
2017-12-11 17:42                             ` Simon Barber
2017-12-01 13:17                           ` [Bloat] " Luca Muscariello
2017-12-01 13:40                             ` Toke Høiland-Jørgensen
2017-12-01 17:42                               ` Dave Taht
2017-12-01 20:39                                 ` Juliusz Chroboczek
2017-12-03  5:20                                   ` [Bloat] [Make-wifi-fast] " Bob McMahon
2017-12-03 10:35                                     ` Juliusz Chroboczek
2017-12-03 11:40                                       ` Jan Ceuleers
2017-12-03 13:57                                         ` Juliusz Chroboczek
2017-12-03 14:07                                           ` Mikael Abrahamsson
2017-12-03 19:53                                             ` Juliusz Chroboczek
2017-12-03 14:09                                           ` Ryan Mounce
2017-12-03 19:54                                             ` Juliusz Chroboczek
2017-12-03 20:14                                               ` Sebastian Moeller
2017-12-03 22:27                                                 ` Dave Taht
2017-12-03 15:25                                           ` Robert Bradley
2017-12-04  3:44                                         ` Dave Taht
2017-12-04 14:38                                           ` David Collier-Brown
2017-12-04 15:44                                             ` Juliusz Chroboczek
2017-12-04 17:17                                               ` David Collier-Brown
2017-12-03 19:04                                       ` Bob McMahon
2017-12-01 21:17                                 ` Bob McMahon
2017-12-01  8:45                       ` [Bloat] " Sebastian Moeller
2017-12-01 10:45                         ` Luca Muscariello
2017-12-01 18:43                           ` Dave Taht
2017-12-01 18:57                             ` Luca Muscariello
2017-12-01 19:36                               ` Dave Taht
2017-11-30 14:51                   ` Neal Cardwell [this message]
2017-11-30 15:55                     ` Eric Dumazet
2017-11-30 15:57                       ` Neal Cardwell
2017-11-29 16:50       ` Sebastian Moeller
2017-12-12 19:27         ` Benjamin Cronce
2017-12-12 20:04           ` Dave Taht
2017-12-12 21:03           ` David Lang
2017-12-12 21:29             ` Jonathan Morton
2017-12-12 22:03           ` Jonathan Morton
2017-12-12 22:21             ` David Lang
     [not found]               ` <CAJq5cE3nfSQP0GCLjp=X0T-iHHgAs=YUCcr34e3ARgkrGZe-wg@mail.gmail.com>
2017-12-12 22:41                 ` Jonathan Morton
2017-12-13  9:46                   ` Mikael Abrahamsson
2017-12-13 10:03                     ` Jonathan Morton
2017-12-13 12:11                       ` Sebastian Moeller
2017-12-13 12:18                         ` Jonathan Morton
2017-12-13 12:36                     ` Sebastian Moeller
2017-12-13 12:39             ` Luca Muscariello
2017-11-29 18:21   ` Juliusz Chroboczek
2017-11-29 18:41     ` Dave Taht
2017-11-29 23:29       ` Steinar H. Gunderson
2017-11-29 23:59       ` Stephen Hemminger
2017-11-30  0:21         ` Eric Dumazet
2017-12-11 20:15   ` Benjamin Cronce
2017-11-29 18:28 ` Juliusz Chroboczek
2017-11-29 18:48   ` Dave Taht
2017-12-11 18:30   ` Jonathan Morton

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

  List information: https://lists.bufferbloat.net/postorius/lists/bloat.lists.bufferbloat.net/

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

  git send-email \
    --in-reply-to='CADVnQyn0x3s6vCg7RtQ=2tWHTNRJDQsm4qu+6PnPAPfv6nyxGw@mail.gmail.com' \
    --to=ncardwell@google.com \
    --cc=bloat@lists.bufferbloat.net \
    --cc=chromatix99@gmail.com \
    --cc=dros@simula.no \
    --cc=eric.dumazet@gmail.com \
    --cc=michawe@ifi.uio.no \
    --cc=weiwan@google.com \
    --cc=ycheng@google.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