[aqm] bailing on aqm

Dave Taht <dave.taht@gmail.com> Fri, 20 March 2015 12:13 UTC

MIME-Version: 1.0
Date: Fri, 20 Mar 2015 05:13:48 -0700
Message-ID: <CAA93jw61uhw5aP=6mjQqX=1_72TBqG0PLxbaroScWq=q=icAzw@mail.gmail.com>
From: Dave Taht <dave.taht@gmail.com>
To: "aqm@ietf.org" <aqm@ietf.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <http://mailarchive.ietf.org/arch/msg/aqm/meipAtzLUOeonMxP7LzN6Gnz-U4>
Subject: [aqm] bailing on aqm
Precedence: list

 I'm sorry, but I do not have any time for ietf on this go-round.

I keep trying to find time to review the remaining documents, and
today was the last day I had free to do so, and only got through one.
I am booked solid, 6 and 7 days a week, for the next 3 weeks, and it's
nearly 5 am today after a 23 hour day.

I tried just now to explain issues with ecn, and all I can suggest is
that you get a black hat
hacker to blow up your assumptions for you.

Pithy summary of the remaining drafts:

codel - there are some tweaks to codel in the queue to be fully
evaluated this summer, but they are second order stuff, only needed as
a stand-alone qdisc.

fq-codel - toke has the pen on the comments thus far, look forward to
a revision by prague. Getting to be a pretty big deployment now...

pie - ship it

docsis-pie ship it

aqm eval guidelines - since nobody writing that has implemented any
code I know of, nor so far as I know, actually run netperf-wrapper
tests and examine output... I'm not going to bother reading it again.
Happy to hear they added a cross traffic test. If there is running
code, point me at it please, offlist.

rfc2309 replacement - ship it. I would like clear language in that
regarding "packet scheduling" vs fair queueing (do a google search for
either phrase, look also on google scholar. I have this problem with
multiple other phrases floating around, and would generally like ietf
drafts to have google-able terms that line up with everything else)

fq - My basic problem with the fq one was that the polemic at the end
needed to go (haven't re-read it), and:

it totally missed the subtle brilliance of van's summary as to what fq
brought to the party.

someone should transcribe all he actually said - I think it was at 38
minutes into the ietf preso or so - I had cut it down to fit on a
slide:

“FQ_Codel provides great isolation... if you've got low rate
videoconferencing and low rate web traffic they never get dropped. A
lot of issues with IW10 go away, because all the other traffic sees is
the front of the queue. You don't know how big its window is, but you
don't care because you are not affected by it. FQ_Codel increases
utilization across your entire networking fabric, especially for
bidirectional traffic...”

“If we're sticking code into boxes to deploy codel, don't do that.
Deploy fq_codel. It's just an across the board win.” - Van Jacobson
ietf-84

page 22 here:

http://www.ietf.org/proceedings/84/slides/slides-84-tsvarea-4.pdf

I am unsubscribing from aqm now, also. Have a great ietf. Good night.

-- 
Dave Täht
Let's make wifi fast, less jittery and reliable again!

https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb

[aqm] bailing on aqm Dave Taht