General list for discussing Bufferbloat
 help / color / mirror / Atom feed
From: David Lang <david@lang.hm>
To: Sebastian Moeller <moeller0@gmx.de>
Cc: bloat <bloat@lists.bufferbloat.net>
Subject: Re: [Bloat] sigcomm wifi
Date: Sun, 31 Aug 2014 15:37:07 -0700 (PDT)	[thread overview]
Message-ID: <alpine.DEB.2.02.1408311536070.23856@nftneq.ynat.uz> (raw)
In-Reply-To: <B5092C96-9556-4BF7-86CA-43F7E9D93C26@gmx.de>

On Mon, 25 Aug 2014, Sebastian Moeller wrote:

> Hi Michael,
>
> On Aug 25, 2014, at 10:01 , Michael Welzl <michawe@ifi.uio.no> wrote:
> [...]
>>
>>> This is a case where a local proxy server can actually make a big difference 
>>> to you. The connections between your mobile devices and the local proxy 
>>> server have a short RTT and so all timeouts can be nice and short, and then 
>>> the proxy deals with the long RTT connections out to the Internet.
>>
>> Adding a proxy to these considerations only complicates them: it's a hard 
>> enough trade-off when we just ask ourselves: how large should a buffer for 
>> the sake of link layer retransmissions be?  (which is closely related to the 
>> question: how often should a link layer try to retransmit before giving up?) 
>> That's what my emails were about. I suspect that we don't have a good answer 
>> to even these questions, and I suspect that we'd better off having something 
>> dynamic than fixed default values.
>
> 	What about framing the retransmissions not in number but rather in time? 
> For example the maximum of either time to transmit a few (say 3?) packet at 
> the current data rate (or maybe one rate lower than current to allow 
> setoriating signal quality) or 20ms (pulled out of thin air, would need some 
> research). The first should make sure we actually retransmit to overcome 
> glitches, and the second should make sure that RTT does not increase to 
> dramatically. This basically assumes that for reasonable interactive traffic 
> we only have a given RTT budget and should make sure not to overspend ;)

Yep, just like BQL helped a lot on the wired side, because it's a good stand-in 
for the time involved, we need to get the same concept through the wifi stack 
and drivers.

David Lang

  parent reply	other threads:[~2014-08-31 22:37 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-19 16:45 Dave Taht
2014-08-20  7:12 ` Eggert, Lars
2014-08-20 14:01   ` Dave Taht
2014-08-20 22:05   ` Jim Gettys
2014-08-21  6:52     ` Eggert, Lars
2014-08-21  7:11       ` Michael Welzl
2014-08-21  8:30         ` David Lang
2014-08-22 23:07           ` Michael Welzl
2014-08-22 23:50             ` David Lang
2014-08-23 19:26               ` Michael Welzl
2014-08-23 23:29                 ` Jonathan Morton
2014-08-23 23:40                   ` Steinar H. Gunderson
2014-08-23 23:49                     ` Jonathan Morton
2014-08-24  1:33                   ` David Lang
2014-08-24  2:29                     ` Jonathan Morton
2014-08-24  5:12                       ` David Lang
2014-08-24  6:26                         ` Jonathan Morton
2014-08-24  8:24                           ` David Lang
2014-08-24  9:20                             ` Jonathan Morton
2014-08-25  7:25                             ` Michael Welzl
2014-08-30  7:20                             ` Jonathan Morton
2014-08-31 20:46                               ` Simon Barber
2014-08-25  7:35                   ` Michael Welzl
2014-08-24  1:09                 ` David Lang
2014-08-25  8:01                   ` Michael Welzl
2014-08-25  8:19                     ` Sebastian Moeller
2014-08-25  8:33                       ` Michael Welzl
2014-08-25  9:18                         ` Alex Burr
2014-08-31 22:37                       ` David Lang [this message]
2014-08-31 23:09                         ` Simon Barber
2014-09-01  0:25                           ` David Lang
2014-09-01  2:14                             ` Simon Barber
2014-08-31 22:35                     ` David Lang
2014-08-21  6:56     ` David Lang
2014-08-21  7:04     ` David Lang
2014-08-21  9:46       ` Jesper Dangaard Brouer
2014-08-21 19:49         ` David Lang
2014-08-21 19:57           ` Steinar H. Gunderson
2014-08-22 17:07             ` Jan Ceuleers
2014-08-22 18:27               ` Steinar H. Gunderson
2014-08-21  8:58     ` Steinar H. Gunderson
2014-08-22 23:34       ` David Lang
2014-08-22 23:41         ` Steinar H. Gunderson
2014-08-22 23:52           ` David Lang
2014-08-22 23:56             ` Steinar H. Gunderson
2014-08-23  0:03               ` Steinar H. Gunderson
2014-08-21  9:23     ` Mikael Abrahamsson
2014-08-21  9:30       ` Steinar H. Gunderson
2014-08-22 23:30         ` David Lang
2014-08-22 23:40           ` Steinar H. Gunderson
2014-08-20  8:30 ` Steinar H. Gunderson
2014-08-21  6:58   ` David Lang
2014-08-24  3:49 Hal Murray
2014-08-24  3:52 ` Jonathan Morton
2014-08-24  5:14 ` David Lang
2014-08-25  7:43   ` Michael Welzl

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=alpine.DEB.2.02.1408311536070.23856@nftneq.ynat.uz \
    --to=david@lang.hm \
    --cc=bloat@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