onion_router stress-testing...

[Bram Cohen]

Roger Dingledine wrote:

> Yes, this makes sense. The problem is that we do too many public key
> ops when we're being flooded with creates. That is, we use almost all of
> each second processing create cells, so we don't process data cells very
> often. Our congestion control algorithm makes it even worse, because
> each circuit can push at most 1200 bytes before needing a sendme cell
> to work its way back through the circuit.

It may make sense to shove the http proxy outside the onion network, and
have it be spoken to by a single TCP connection. That would get rid of all
the creates when web browsing.

By the way, the new CodeCon CFP is out, submissions are due December 15.

-Bram