[e2e] Can feedback be generated more fast in ECN?

[Francesco Potorti`]

     SQ only requires a single message on the uncongested leg, rather
     than a round-trip through the known congested leg.

ECN  should be  used  in conjunction  with  RED, having  the ability  of
signaling incumbent  congestion without adding  any more traffic  to the
network.  You cannot do the same with SQ.

You could object that in case of incumbent congestion you shouldn't have
a problem sending  some packets on the uncongested leg  of the path, but
others have signaled how this could be expensive for the routers.

Maybe  the  two mechanisms  could  coexist  and  be used  for  different
purposes, ECN  for incumbent  congestion, SQ when  a packet  is dropped,
provided the  latter occurrency  is much less  frequent than  the former
one.  In that view, they would be almost orthogonal to each other.
   
     SQ works with UDP

So can ECN.  The two mechanisms are both at the same level in the stack.
   
     SQ can work in conjunction with streaming and interactive
     applications equally (knock down codec based on SQ code point
     value for example, or increase SRTT as another).

When congestion control is receiver based ECN should be the choice.

-- 
Francesco Potortì (researcher)         Voice: +39 050 315 3058 (op.2111)
Area della ricerca CNR - CNUCE         Fax:   +39 050 313 8091
via Alfieri 1, I-56010 Ghezzano, Pisa  Email: F.Potorti at cnuce.cnr.it
Web: http://fly.cnuce.cnr.it/          Key:   fly.cnuce.cnr.it/public.key