Date: Fri, 31 Oct 2008 13:24:39 +0200
From: Lars Eggert <lars(_dot_)eggert(_at_)nokia(_dot_)com>
Message-ID: <609C5AC4-D360-43E7-90CD-B33A3265E4E2(_at_)nokia(_dot_)com>
| > The first bullet says "deal with the world as it is"; the second
| > says "deal with the world as you wish it were"
| >
| > I think that is a very sensible approach.
Well, yes, but only if you really believe that the first bullet is
"the world as it is". As it mostly is, probably, but completely?
I'm just a little concerned about routers that currently classify packets
based upon protocol/address/port info (ignoring TOS/DSCP) and then
don't do FIFO queueing. But aren't "as you wish it were" either.
I didn't see anything in the charter that covered that intermediate case.
I don't necessarily expect that everything would "just work" with
such a router, but I'd like to be assured that updating its classification
engine (ie: adding a few new rules) will do the right thing.
| It could be easily distinguishable (for example, we could decide to
| use the LTBE DSCP for all packets), but it doesn't need to be. LEDBAT
| is doing an end-to-end algorithm that is intended to work without any
| router support.
Yes, I understood that - what I'm concerned about is not (completely)
breaking routers that are currently trying to do something better than
nothing (where "breaking" means having background traffic given higher
priority than any other traffic.)
| Right, the intent is that the LEDBAT mechanism should work at least as
| well if some sort of network help is available than when there isn't.
| Maybe we should make that explicit.
That is exactly what I was suggesting.
kre
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf