[Iccrg] About cautious use of unreliable feedback

[Lachlan Andrew]

Greetings Cesar,

On 03/02/2008, Cesar Marcondes <cmarcond06 at gmail.com> wrote:
> Finally, one can argue that, the router changes happen
> on another time granularity larger than the lifetime of a typical
> (HTTP) TCP connection.

You can argue that, but the algorithm should "work" in all cases, not
just the typical cases.  (By "work", I don't mean "work as intended",
but "not fail miserably".  CTCP doesn't work as intended if the
baseRTT increases without a timeout, but by returning to Reno, it
returns to something not too broken.)

Cheers,
Lachlan

-- 
Lachlan Andrew  Dept of Computer Science, Caltech
1200 E California Blvd, Mail Code 256-80, Pasadena CA 91125, USA
Ph: +1 (626) 395-8820    Fax: +1 (626) 568-3603
http://netlab.caltech.edu/~lachlan