[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: Squeezing non-relays at the entry node



Thus spake Roger Dingledine (arma@xxxxxxx):

> Hi folks (Nick in particular),
> 
> I've been pondering other performance improvements. One of them is to
> rate-limit client connections as they enter the network. Rate limiting
> in the Tor client itself would work better, but it's not a very stable
> equilibrium -- it encourages people to switch to security disasters
> like tortunnel.
>
> It would impact Mike's bwauthority tests. We'd want to make an exception
> for those Tors. I think we'd leave the torperf deployments alone, since
> after all their goal is to measure "realistic" client performance.

Has code been checked in to handle this case? Otherwise any
experiments you run will be tainted by the bias of the bandwidth
authorites effiectively being disabled, for better or worse.

Also, my opinion is that eventually this value should be set by the
bandwidth authorites. They already measure a network-wide average
stream capacity. It seems to me that this should be the value we cap
long term client streams at, if it turns out that this experiment
improves overall performance. 

It does feel like a hack to compensate for poor flow control to me,
though.


-- 
Mike Perry
Mad Computer Scientist
fscked.org evil labs

Attachment: pgpuH4g4coup9.pgp
Description: PGP signature