[Iccrg] Proposal for ICCRG operations

[Saverio Mascolo]

> I think you are assuming that the queue occupancy for each flow is
> constant over time.  If you relax that assumption, then you no longer
> have the implication.
> 
> There may exist viable control schemes which control the flows (and
> have whatever other desirables properties) while keeping the average
> queue occupancy small.
> 
> Do you agree?

then i agree and is  necessary to decide how to fix the queue dynamics. 

> 
> > i have read somewhere, maybe on s. floud homepage, a note on "why
> > control queue length should not be a target". i am not sure, but i think
> > she had  similar arguments.
> 
> I looked on her web pages at http://www.icir.org/floyd/papers.html
> and  http://www.icir.org/floyd/notes.html  and did not manage to
> identify which note you would be refering to.

i am sure it was a note written by floyd. maybe she can help on this.

saverio

> 
> > in other words, if you control queue at fixed value you would prevent
> > other flows to join.
> 
> I think there must be assumptions in your reasoning that I did not
> mean to imply.
> 
> 			-Tim Shepard
> 			 shep at alum.mit.edu