From: Jonathan Morton <chromatix99@gmail.com>
To: Sebastian Moeller <moeller0@gmx.de>
Cc: "Dave Täht" <dave.taht@gmail.com>,
bloat <bloat@lists.bufferbloat.net>,
ecn-sane@lists.bufferbloat.net
Subject: Re: [Ecn-sane] [Bloat] [iccrg] Fwd: [tcpPrague] Implementation and experimentation of TCP Prague/L4S hackaton at IETF104
Date: Fri, 15 Mar 2019 17:49:34 +0200 [thread overview]
Message-ID: <D59E639C-24A6-44F4-94A4-CBD66209B42D@gmail.com> (raw)
In-Reply-To: <CA59C597-4EBC-46E7-BD73-24037273107F@gmx.de>
> On 15 Mar, 2019, at 4:44 pm, Sebastian Moeller <moeller0@gmx.de> wrote:
>
> In essence, they do not want to deal with the diffserv mess under the end-to-end perspective and making it reliable.
Yeah, that's pretty much what I thought. Diffserv really does need to be fixed sometime.
>> But Codel-type AQMs don't provide the ideal ECN signal for L4S (and nor do RED-type AQMs without specific tuning for L4S), and any single-queue AQM is going to end up with L4S flows outcompeting standard ones quite seriously.
>
> Except L$S flows are required to "degrade" to classic congestion contro once they have proof of not being handled by a l4s aware AQM, like real packet drops or ECT(0) + CE...
That isn't enough, because ECN AQMs as presently specified won't change ECT(1) to ECT(0) (nor will SCE), and it would require a lot of overload before dropping actually started. So those signals don't reliably identify a legacy AQM.
> L4S would find uses for SCE, but that still faces the same roll-out issue...
It's not the same roll-out issue, because in an SCE context L4S would have to be legacy-compatible by default, and only employ the "new" behaviour when SCE information appears - the reverse of its present behaviour - which then makes it backwards compatible and incrementally deployable.
The real snag is that DCTCP's setpoint is 2 marks per RTT, which is different from SCE's specified setpoint of 50% packets marked (unless the cwnd is down to 4 packets). That'd make it difficult for a straight port of DCTCP to SCE to achieve full throughput.
A sane compromise could be for SCE to be marked in the same way as L4S marks CE, and a valid interpretation of SCE to follow the 1/n response of DCTCP. That would leverage existing TCP-Prague R&D, but in a backwards-compatible, incrementally-deployable manner. Perhaps that's something worth discussing at IETF?
- Jonathan Morton
next prev parent reply other threads:[~2019-03-15 15:49 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <AM0PR07MB48198660539171737E4CCAB1E0730@AM0PR07MB4819.eurprd07.prod.outlook.com>
[not found] ` <d91a6a71-5898-9571-2a02-0d9d83839615@bobbriscoe.net>
2019-03-15 10:46 ` [Ecn-sane] " Dave Taht
[not found] ` <1E80578D-A589-4CA0-9015-B03B63042355@gmx.de>
2019-03-15 14:06 ` [Ecn-sane] [Bloat] " Dave Taht
2019-03-15 15:52 ` Sebastian Moeller
2019-03-15 17:01 ` David P. Reed
2019-03-15 17:45 ` Sebastian Moeller
2019-03-15 18:36 ` Mikael Abrahamsson
2019-03-15 19:23 ` Sebastian Moeller
2019-03-15 19:32 ` Jonathan Morton
2019-03-15 19:44 ` David P. Reed
2019-03-15 20:13 ` Jonathan Morton
2019-03-15 23:43 ` David P. Reed
2019-03-16 1:26 ` Jonathan Morton
2019-03-16 7:38 ` Sebastian Moeller
2019-03-16 18:56 ` Michael Richardson
2019-03-15 20:28 ` Jonathan Foulkes
2019-03-15 20:31 ` Dave Taht
2019-03-15 23:45 ` David P. Reed
2019-03-16 9:42 ` Michael Welzl
2019-03-16 10:08 ` Sebastian Moeller
2019-03-16 10:23 ` Nils Andreas Svee
2019-03-16 14:55 ` Jonathan Foulkes
2019-03-16 21:38 ` Holland, Jake
2019-03-16 21:57 ` Vint Cerf
2019-03-16 22:03 ` Dave Taht
2019-03-16 22:05 ` Holland, Jake
2019-03-17 18:07 ` David P. Reed
2019-03-17 18:05 ` Vint Cerf
2019-03-19 4:44 ` Greg White
2019-03-19 5:35 ` Jonathan Morton
2019-03-19 5:52 ` Greg White
2019-03-19 7:10 ` Jonathan Morton
2019-03-19 8:07 ` Sebastian Moeller
2019-03-19 8:50 ` Sebastian Moeller
2019-03-19 23:59 ` Dave Taht
2019-03-20 10:17 ` Sebastian Moeller
[not found] ` <5458c216-07b9-5b06-a381-326de49b53e0@bobbriscoe.net>
[not found] ` <AC14ACBB-A7CC-40E0-882C-2519D05ADC05@akamai.com>
[not found] ` <5C9296E1.4010703@erg.abdn.ac.uk>
[not found] ` <F62C4839-0489-475F-AD8F-58913EEEEC0F@gmail.com>
[not found] ` <FDA48F4C-415B-4B8E-9CC7-2AAAD4DC3BE8@cablelabs.com>
2019-03-20 22:12 ` [Ecn-sane] [Bloat] [tsvwg] " Sebastian Moeller
2019-03-20 22:31 ` Jonathan Morton
2019-03-20 22:56 ` Sebastian Moeller
2019-03-20 23:03 ` Jonathan Morton
2019-03-20 23:11 ` Holland, Jake
2019-03-20 23:28 ` Jonathan Morton
2019-03-21 8:15 ` Mikael Abrahamsson
2019-03-21 8:31 ` Mikael Abrahamsson
2019-03-20 23:30 ` Sebastian Moeller
2019-03-21 0:15 ` Holland, Jake
2019-03-16 22:03 ` [Ecn-sane] [Bloat] " Jonathan Morton
2019-03-16 22:09 ` Sebastian Moeller
2019-03-17 14:06 ` Mikael Abrahamsson
2019-03-17 17:37 ` Loganaden Velvindron
2019-03-17 17:40 ` Toke Høiland-Jørgensen
2019-03-17 17:44 ` Mikael Abrahamsson
2019-03-17 18:00 ` Dave Taht
2019-03-17 19:38 ` Rodney W. Grimes
2019-03-17 20:50 ` Luca Muscariello
2019-03-17 21:51 ` Toke Høiland-Jørgensen
2019-03-18 4:26 ` Mikael Abrahamsson
2019-03-16 4:04 ` Jonathan Morton
2019-03-16 4:51 ` Dave Taht
2019-03-15 18:07 ` Mikael Abrahamsson
2019-03-15 14:27 ` Jonathan Morton
2019-03-15 14:44 ` Sebastian Moeller
2019-03-15 15:49 ` Jonathan Morton [this message]
2019-03-15 21:34 ` Wesley Eddy
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/ecn-sane.lists.bufferbloat.net/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=D59E639C-24A6-44F4-94A4-CBD66209B42D@gmail.com \
--to=chromatix99@gmail.com \
--cc=bloat@lists.bufferbloat.net \
--cc=dave.taht@gmail.com \
--cc=ecn-sane@lists.bufferbloat.net \
--cc=moeller0@gmx.de \
/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