From: MUSCARIELLO Luca IMT/OLN <luca.muscariello@orange.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: bloat <bloat@lists.bufferbloat.net>
Subject: Re: [Bloat] Pacing --- was DSLReports Speed Test has latency measurement built-in
Date: Wed, 22 Apr 2015 20:39:20 +0200 [thread overview]
Message-ID: <5537EAD8.3060806@orange.com> (raw)
In-Reply-To: <1429726944.18561.122.camel@edumazet-glaptop2.roam.corp.google.com>
This is not clear to me in general.
I can understand that for the first shot of the IW>=10 pacing is always
a win strategy no matter what queuing system you have
because it reduces the loss probability in that window. Still fq_codel
would reduce that probability even more.
But for long flows going far beyond that phase isn't clear to me why the
paced flow is not penalized by the non paced flow in FIFO.
TCP will start filling the pipe at some point in bursts and that would hurt.
Now, I forgot how sch_fq pacing rate is initialized to be effective from
the very first window.
On 04/22/2015 08:22 PM, Eric Dumazet wrote:
> It really does.
>
> This is why we deployed sch_fq and let our competitors find this later.
next prev parent reply other threads:[~2015-04-22 18:39 UTC|newest]
Thread overview: 127+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-04-19 5:26 [Bloat] " jb
2015-04-19 7:36 ` David Lang
2015-04-19 7:48 ` David Lang
2015-04-19 9:33 ` jb
2015-04-19 10:45 ` David Lang
2015-04-19 8:28 ` Alex Burr
2015-04-19 10:20 ` Sebastian Moeller
2015-04-19 10:46 ` Jonathan Morton
2015-04-19 16:30 ` Sebastian Moeller
2015-04-19 17:41 ` Jonathan Morton
2015-04-19 19:40 ` Sebastian Moeller
2015-04-19 20:53 ` Jonathan Morton
2015-04-21 2:56 ` Simon Barber
2015-04-21 4:15 ` jb
2015-04-21 4:47 ` David Lang
2015-04-21 7:35 ` jb
2015-04-21 9:14 ` Steinar H. Gunderson
2015-04-21 14:20 ` David Lang
2015-04-21 14:25 ` David Lang
2015-04-21 14:28 ` David Lang
2015-04-21 22:13 ` jb
2015-04-21 22:39 ` Aaron Wood
2015-04-21 23:17 ` jb
2015-04-22 2:14 ` Simon Barber
2015-04-22 2:56 ` jb
2015-04-22 14:32 ` Simon Barber
2015-04-22 17:35 ` David Lang
2015-04-23 1:37 ` Simon Barber
2015-04-24 16:54 ` David Lang
2015-04-24 17:00 ` Rick Jones
2015-04-21 9:37 ` Jonathan Morton
2015-04-21 10:35 ` jb
2015-04-22 4:04 ` Steinar H. Gunderson
2015-04-22 4:28 ` Eric Dumazet
2015-04-22 8:51 ` [Bloat] RE : " luca.muscariello
2015-04-22 12:02 ` jb
2015-04-22 13:08 ` Jonathan Morton
[not found] ` <14ce17a7810.27f7.e972a4f4d859b00521b2b659602cb2f9@superduper.net>
2015-04-22 14:15 ` Simon Barber
2015-04-22 13:50 ` [Bloat] " Eric Dumazet
2015-04-22 14:09 ` Steinar H. Gunderson
2015-04-22 15:26 ` [Bloat] RE : " luca.muscariello
2015-04-22 15:44 ` [Bloat] " Eric Dumazet
2015-04-22 16:35 ` MUSCARIELLO Luca IMT/OLN
2015-04-22 17:16 ` Eric Dumazet
2015-04-22 17:24 ` Steinar H. Gunderson
2015-04-22 17:28 ` MUSCARIELLO Luca IMT/OLN
2015-04-22 17:45 ` MUSCARIELLO Luca IMT/OLN
2015-04-23 5:27 ` MUSCARIELLO Luca IMT/OLN
2015-04-23 6:48 ` Eric Dumazet
[not found] ` <CAH3Ss96VwE_fWNMOMOY4AgaEnVFtCP3rPDHSudOcHxckSDNMqQ@mail.gmail.com>
2015-04-23 10:08 ` jb
2015-04-24 8:18 ` Sebastian Moeller
2015-04-24 8:29 ` Toke Høiland-Jørgensen
2015-04-24 8:55 ` Sebastian Moeller
2015-04-24 9:02 ` Toke Høiland-Jørgensen
2015-04-24 13:32 ` jb
2015-04-24 13:58 ` Toke Høiland-Jørgensen
2015-04-24 16:51 ` David Lang
2015-04-25 3:15 ` Simon Barber
2015-04-25 4:04 ` Dave Taht
2015-04-25 4:26 ` Simon Barber
2015-04-25 6:03 ` Sebastian Moeller
2015-04-27 16:39 ` Dave Taht
2015-04-28 7:18 ` Sebastian Moeller
2015-04-28 8:01 ` David Lang
2015-04-28 8:19 ` Toke Høiland-Jørgensen
2015-04-28 15:42 ` David Lang
2015-04-28 8:38 ` Sebastian Moeller
2015-04-28 12:09 ` Rich Brown
2015-04-28 15:26 ` David Lang
2015-04-28 15:39 ` David Lang
2015-04-28 11:04 ` Mikael Abrahamsson
2015-04-28 11:49 ` Sebastian Moeller
2015-04-28 12:24 ` Mikael Abrahamsson
2015-04-28 13:44 ` Sebastian Moeller
2015-04-28 19:09 ` Rick Jones
2015-04-28 14:06 ` Dave Taht
2015-04-28 14:02 ` Dave Taht
2015-05-06 5:08 ` Simon Barber
2015-05-06 8:50 ` Sebastian Moeller
2015-05-06 15:30 ` Jim Gettys
2015-05-06 18:03 ` Sebastian Moeller
2015-05-06 20:25 ` Jonathan Morton
2015-05-06 20:43 ` Toke Høiland-Jørgensen
2015-05-07 7:33 ` Sebastian Moeller
2015-05-07 4:29 ` Mikael Abrahamsson
2015-05-07 7:08 ` jb
2015-05-07 7:18 ` Jonathan Morton
2015-05-07 7:24 ` Mikael Abrahamsson
2015-05-07 7:40 ` Sebastian Moeller
2015-05-07 9:16 ` Mikael Abrahamsson
2015-05-07 10:44 ` jb
2015-05-07 11:36 ` Sebastian Moeller
2015-05-07 11:44 ` Mikael Abrahamsson
2015-05-07 13:10 ` Jim Gettys
2015-05-07 13:18 ` Mikael Abrahamsson
2015-05-07 13:14 ` jb
2015-05-07 13:26 ` Neil Davies
2015-05-07 14:45 ` Simon Barber
2015-05-07 22:27 ` Dave Taht
2015-05-07 22:45 ` Dave Taht
2015-05-07 23:09 ` Dave Taht
2015-05-08 2:05 ` jb
2015-05-08 4:16 ` David Lang
2015-05-08 3:54 ` Eric Dumazet
2015-05-08 4:20 ` Dave Taht
2015-05-08 13:20 ` [Bloat] DSLReports Jitter/PDV test Rich Brown
2015-05-08 14:22 ` jb
2015-05-07 7:37 ` [Bloat] DSLReports Speed Test has latency measurement built-in Sebastian Moeller
2015-05-07 7:19 ` Mikael Abrahamsson
2015-05-07 6:19 ` Sebastian Moeller
2015-04-25 3:23 ` Simon Barber
2015-04-24 15:20 ` Bill Ver Steeg (versteb)
2015-04-25 2:24 ` Simon Barber
2015-04-23 10:17 ` renaud sallantin
2015-04-23 14:10 ` Eric Dumazet
2015-04-23 14:38 ` renaud sallantin
2015-04-23 15:52 ` Jonathan Morton
2015-04-23 16:00 ` Simon Barber
2015-04-23 13:17 ` MUSCARIELLO Luca IMT/OLN
2015-04-22 18:22 ` Eric Dumazet
2015-04-22 18:39 ` MUSCARIELLO Luca IMT/OLN [this message]
2015-04-22 19:05 ` [Bloat] Pacing --- was " Jonathan Morton
2015-04-22 15:59 ` [Bloat] RE : " Steinar H. Gunderson
2015-04-22 16:16 ` Eric Dumazet
2015-04-22 16:19 ` Dave Taht
2015-04-22 17:15 ` Rick Jones
2015-04-19 12:14 ` [Bloat] " Toke Høiland-Jørgensen
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=5537EAD8.3060806@orange.com \
--to=luca.muscariello@orange.com \
--cc=bloat@lists.bufferbloat.net \
--cc=eric.dumazet@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