Re: [dccp] question about gaming congestion control

[Eddie Kohler <kohler@icir.org>]

> One more thing: Rob wondered if any attention had been payed to people
> "gaming" the congestion control algorithms, i.e., similar to the
> ack-splitting attack of a few years ago.

We've thought about that throughout the protocol. Ack-splitting is less of
an issue here, with per-packet sequence numbers. We will have text in the
CCIDs about acking stuff that hasn't been sent. Our basic mechanism for
preventing sender misbehavior is ECN Nonce verification (the sender can
probabilistically verify that the receiver really did receive non-marked
every packet it said it received non-marked). DCCP ubiquitously supports
nonce verification -- we changed our design for several features, such as
Data Dropped and TFRC's loss rate reporting, to facilitate it. Of course we
might have missed something.

Eddie


_______________________________________________
dccp IETF mailing list: dccp@ietf.org
list info:  https://www1.ietf.org/mailman/listinfo/dccp
wg charter: http://www.ietf.org/html.charters/dccp-charter.html