[e2e] Internet packet dynamics

[Armando L. Caro Jr.]

On Sun, 14 Mar 2004, Sam Liang wrote:

>   I think it will be very important to get data on the percentage of each
> cause of packet loss. It's very important to distinguish the cause of the
> packet loss, because you want to respond differently.  Loss due to outages
> should not trigger congestion backoff (assuing such outage is not caused
> by congestion), just like packet loss in the wireless environment.

First, I don't see how it is possible to distinguish congestion induced
outages from non-congestion induced outages. Maybe it's possible in some
scenarios if the sender tracks delay variations similar to how FAST TCP
does, but this technique probably won't work with flash crowds.

Second, I am not convinced that senders shouldn't back off when noisy
wireless environments cause loss. If the wireless link layer does
retransmissions, noisy loss will also cause congestion and the sender
should back off.

~armando

0--                                              --0
| Armando L. Caro Jr.  |  Protocol Engineering Lab |
| www.armandocaro.net  |    University of Delaware |
0--                                              --0