[e2e] pre-BOF discussion: "ABR to the Internet"

[Douglas Otis]

All,

Although today's CPUs are capable of indicating elapse time at the CPU clock
resolution using a single instruction, the OS and networking interface
depends on much lower time resolutions.  As such, setting rates versus the
self pacing scheme will demand that a high resolution of time be made
available and that transmissions can be scheduled based on rate intervals.
This would seem to be a major departure from the present techniques.
Available Bit Rate (ABR) could translate into a send cap of some type, but
it will be difficult to ensure any actual use without a higher time
resolution than typically available to the drivers in my opinion.

Doug

> Michael,
>
> my work on the FS-paradigm and on PLM show, among other things, that an
> explicit
> rate notification (based on Packet Pairs, thus without overhead) allows
> for a dramatically more efficient congestion control protocol.
>
>
http://www.eurecom.fr/~btroup/FormerPages/legout_htdoc/Research/research.htm
l
>
>
> I believe that finding a simple mechanism for explicit rate notification
> will
> definitely simplify the problem of congestion control. The main problem
> is how to define such a mechanism, the design of congestion control
> protocols with such a network support is very easier than with
> the current FIFO networks.
>
> I think that working toward this direction will be highly interesting,
> but
> highly controversial as I experience during my PhD. Moreover the
> practical
> deployment of such an explicit rate based notification will be very
> difficult,
> as it will probably require significant modifications of the network.
>
>
> Regards,
> Arnaud.
>
>
>
>
> Michael Welzl wrote:
> >
> > Hello everyone,
> >
> > I would like to start a discussion to gauge interest in possibly having
> > a BOF (note that nothing has been decided yet - but starting with
> > a mailing list and a web site seems to be a good idea) at the
> London IETF
> > meeting.
> >
> > The topic is "ABR to the Internet" - in fact, it's "ABR Explicit Rate
> > Feedback related mechanisms for the Internet". To quote from
> the web site:
> >
> > The goal: A generic approach to ...
> > - identify relevant information
> > - propose a new protocol and/or changes to existing protocols
> > - propose ways to use specific information (endpoint behaviour)
> >   to achieve a certain QoS gain
> > Don't think of it as strictly ABR-like (requiring per flow state).
> > Think of it as ECN with a finer granularity.
> >
> > Let us see if we can find enough common ground to justify a BOF request.
> > To join the discussion, subscribe to the mailing list at:
> > http://www.tk.uni-linz.ac.at/~michael/abr-internet/
> > You will find about anything you need at this web site.
> >
> > Kind regards,
> > Michael Welzl
> > ----------------------------------------------------------------------
> > Dipl.-Ing. Michael Welzl     Researcher
> > Telecooperation Group        Dpt. of Computer Science
> > University of Linz           Altenberger Str. 69, A-4040 Linz, Austria
> > Phone: +43/732/2468-9264     Fax: +43/732/2468-9829
> > ICQ # 38201212               http://www.tk.uni-linz.ac.at/~michael/
>
> --
> ----------------------------------------------------------------------
> Arnaud Legout, Ph.D.
>
> Castify Networks                Phone : 00.33.4.92.94.20.91
> 2229, route des Cretes          Fax   : 00.33.4.92.94.20.88
> 06560 Sophia Antipolis          E-mail: legout at castify.net
> FRANCE                          Web   : http://www.castify.net
> ----------------------------------------------------------------------
>