On 3/7/2014 12:20 AM, Matt Mathis wrote:
The ongoing debate about tunnels and congestion control is over the
wrong question. The presence of congestion control in the tunnel is
neither sufficient nor necessary to protect the network from the tunnel.
Let me define two terms: traffic or applications are "elastic" if
increasing loss causes decreasing presented load and "regenerative" if
increasing loss causes increased load (from the sender to the loss
point).
Now you've completely lost me.
If congestion isn't increased loss due to increased load (and then CC
isn't what you call "elastic"), what is?
I agree that TCP's CC can be at odds with its error recovery, and that
this can be amplified if there are lots of flows that share a path, but
that's not saying that CC failed IMO. Just that TCP doesn't to the best
job of prioritizing CC over error recovery in some environments.
Joe
