Tony, draft-ietf-isis-ext-eth-01 predates the August 2005 report by several years. This thing comes back every few years, and each time we get a little further. (We being the pro-jumbo camp).
I have most recently asked for something that should be straight forward: would the 802.3 vendors who claim to support jumbo please define what that means, in a way that supports interoperability? This sounds trivial, but is not, because there are a large number ways of implementing Ethernet that have implicit MTU limits. And not too surprisingly, there are also know examples of gear that only sort of delivers jumbo packets, or that does not interoperate with other vendors. In a simplistic view, such a document needs to say that the "clock rate adaptation" is implemented by doing frame level store and forward, not a bit or byte level FIFO. I.e. The receiver needs to be able to collect an entire frame at the maximum supported MTU even under the permitted worst case tolerance for the sender and receiver clock rates. It should also require a settable, enforced MTU limit, so an network operator can choose the balance between jitter, buffer resources and MTU. All of the reputable vendors' products do this already. But there are some who don't. Thanks, --MM-- ------------------------------------------- Matt Mathis http://www.psc.edu/~mathis Work:412.268.3319 Home/Cell:412.654.7529 ------------------------------------------- Evil is defined by mortals who think they know "The Truth" and use force to apply it to others. On Mon, 27 Aug 2007, Tony Li wrote: > > On Aug 27, 2007, at 6:16 AM, Bernard Aboba wrote: > > >>> Here are the minutes from a discussion with IEEE 802.3 on the > >>> subject of Jumbo Frames in August 2005: > >>> http://www.iab.org/liaisons/ieee/2005-08-ieee802-liaison-report.html > >> > >> Did anything happen in the IETF as a result of this? > > > > As far as I know, work on a Jumbo Frame document did not go forward > > within IETF. However, I assume that the offer of assistance from > > IEEE 802.3 still remains valid. > > > To the best of my recollection, the work on Jumbo Frames culminated > in http://tools.ietf.org/html/draft-ietf-isis-ext-eth-01 (and > possibly one other draft that's now lost in the mists of time) and > was primarily driven by one particular operator. Given personnel > turnover at the operator, I think that you'll find it hard to find a > motivated constituency, especially given the chilly reception the > first time. > > Tony > > > _______________________________________________ > Int-area mailing list > [email protected] > https://www1.ietf.org/mailman/listinfo/int-area > _______________________________________________ Int-area mailing list [email protected] https://www1.ietf.org/mailman/listinfo/int-area
