RE: Jumbo Frames in 10GbE?
- To: "'Thirion, Walt '" <wthirion@xxxxxxxxxx>, "''stds-802-3-hssg@xxxxxxxx' '" <stds-802-3-hssg@xxxxxxxx>
- Subject: RE: Jumbo Frames in 10GbE?
- From: Andrew Smith <andrew@xxxxxxxxxxxxxxxxxxx>
- Date: Tue, 22 Jun 1999 00:00:35 -0700
- Sender: owner-stds-802-3-hssg@xxxxxxxxxxxxxxxxxx
Just as a point of information, the "IETF document" in question is a private
individual's submission as an "internet draft" - it has no standing as any
type of IETF standards' document at this time. It will expire in December
1999 if no further action is taken.
Anyone can submit internet drafts at any time for anyone (or, often, no-one)
to read. As far as I am aware there is no endorsement of this draft by any
of the IETF's working groups, by the Internet Engineering Steering Group
(kind of like the IEEE 802 Exec) or by the Internet Architecture Board. Of
course that might change. It might be a good idea for one of the draft's
authors to step forward and explain to this group whether they have any
intentions of asking any of the IETF's official bodies to act further on
From: Thirion, Walt
Sent: 6/21/99 10:28 PM
Subject: RE: Jumbo Frames in 10GbE?
Just because there is an IETF document that implementers are free to
follow doesn't mean the base IEEE standard should be changed. Expanding
the frame size can cause major incompatibility with previous
implementations and that, in my opinion, would obviate one of the very
reasons that ethernet continues to be successful.
Moreover, segmentation can be done with a smart network interface,
thereby offloading the per packet load on the host without breaking the
This does not mean, however, that hardware shouldn't be built that can
handle larger packets. As I said, that is left to the implementer. If a
network designer wants to turn that feature "on", he can do so at his
own risk AND with the understanding it isn't a standard setup.
There is some precedent for this in the early days of ethernet in that
many PHYs were capable of driving significantly longer cables than
specified in the standard. Implementers were free to take advantage of
this when they felt it had value. But, because it was non-standard,
there was no guarantee that changing a piece of equipment on either end
of the wire would still work over the longer cables. But it was their
Level One Communications
> -----Original Message-----
> From: owner-stds-802-3-hssg@xxxxxxxxxxxxxxxxxx
> [ mailto:owner-stds-802-3-hssg@xxxxxxxxxxxxxxxxxx
<mailto:owner-stds-802-3-hssg@xxxxxxxxxxxxxxxxxx> ]On Behalf Of
> Joe Gwinn
> Sent: Monday, June 21, 1999 2:39 PM
> To: Booth, Brad
> Cc: HSSG_reflector (E-mail)
> Subject: Jumbo Frames in 10GbE?
> At 4:24 PM 99/6/17, Booth, Brad wrote:
> >Just a small point. One of the objectives that passed with
> greater than
> >75% in Coeur d'Alene was to "preserve minimum and maximum
> FrameSize of
> >current 802.3 Std."
> I don't know that the issue is going to stay decided all that
> long, based
> on the recent article "Jumbo Frames gather support" (Jeff
> Caruso, Network
> World, 14 June 1999, page 6), which states that IETF has published a
> working document proposing that ethernet frames be made
> larger than the
> current 1,500-byte maximum, the basic rationale being to
> reduce the packet
> rate and thus load on packet-handling components of the
> system. In short,
> this is a system issue, and cannot really be decided solely at the MAC
> If jumbo frames are to come, 10GbE would be a logical place to start.
> The issue will ultimately be decided by an IEEE Ballot Group, not a
> Plenary. If the market is really going to bigger packets, as
> this article
> implies, it will be hard to resist.
> Joe Gwinn
> The above is in response to the following:
> > >Issues 3 - Bit Error Rate
> > >The assumption will be that this is 10-12. If
> someone wishes to
> > >this they should bring a presentation to the next
> meeting providing
> > detailed reasoning why this needs to change.
> > It strikes me that the issue of larger maximum
> packet sizes will likely
> > come up, just as it did for GbE. If 10GbE goes to 9
> KB packets,
> >the design
> > center BER would need to go to 10^-13 to maintain
> the same theoretical
> > packet loss rate. I'm not sure how much effect
> this would have in
> > practice, as most gigabit links achieve much better
> than 10^-12,
> >if they
> > work at all. Anyway, these items are ripe for
> debate and decision.
> **** end of message ****