W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2014

Re: #540: "jumbo" frames

From: Roberto Peon <grmocg@gmail.com>
Date: Wed, 25 Jun 2014 14:21:42 -0700
Message-ID: <CAP+FsNfP-ZpRYh7o6R4_hEj-MRxunHsTEq2JHrcGWA1RUB8zBQ@mail.gmail.com>
To: "K.Morgan@iaea.org" <K.Morgan@iaea.org>
Cc: Poul-Henning Kamp <phk@phk.freebsd.dk>, Mike Belshe <mike@belshe.com>, Patrick McManus <pmcmanus@mozilla.com>, Mark Nottingham <mnot@mnot.net>, Willy Tarreau <w@1wt.eu>, Greg Wilkins <gregw@intalio.com>, HTTP Working Group <ietf-http-wg@w3.org>, Martin Dürst <duerst@it.aoyama.ac.jp>
On Wed, Jun 25, 2014 at 2:08 PM, <K.Morgan@iaea.org> wrote:

> On 25 June 2014 22:26, grmocg@gmail.com wrote:
> > I'll note that I work at YouTube, and care deeply about video
> distribution... and I'm not worried about it.
> > As a matter of fact, I'd really, really like to take advantage of the
> prioritization scheme along with smaller
> > frames so that I can have better codecs in the future which do much
> better adaptive stuff w.r.t. bandwidth
> > changes.
> > This matters far more to me.
>
> And nothing about the jumbo frames proposal is stopping you.
>

Assuming I stick lots of END_SEGMENTs in there, you might be right.
If I don't do that, however, a proxy might reframe into larger than 16k
frames and screw with the user experience.
Other servers are not likely to be as smart as our servers (as has been
demonstrated in the past), and so user experience with those may suffer.
If there are enough poor implementations, second order effects begin to
affect our usecases as they effect the browsers.


>
>
> > As it turns out, our users care deeply about the quality of their
> experience.
>
> This is such a ridiculous snark I'm at a loss how to even respond.
>

I'm sorry you interpreted it that way, but this consideration drives a
substantial portion of what I do.

Jumbo frames are unlikely to offer any additional quality of experience for
our users, and would not noticeably decrease CPU consumption, despite
supposedly being a motivating usecase.
As a result, jumbo frames are uninteresting for the YouTube usecase.

-=R



>
> This email message is intended only for the use of the named recipient.
> Information contained in this email message and its attachments may be
> privileged, confidential and protected from disclosure. If you are not the
> intended recipient, please do not read, copy, use or disclose this
> communication to others. Also please notify the sender by replying to this
> message and then delete it from your system.
>
Received on Wednesday, 25 June 2014 21:22:09 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:14:31 UTC