Re: udp transport PoC

[Nick Mathewson <nickm@xxxxxxxxxxxxx>]

On Tue, May 13, 2008 at 10:31:21PM -0700, coderman wrote:
> On Tue, May 13, 2008 at 3:11 PM, Adam Langley <agl@xxxxxxxxxxxxxxxxxx> wrote:
 [...]
> >  Your flow control should probably be based on[2]. The description in
> >  the TCP RFC now bares no relation to reality.
> 
> the problem with trying for true tcp fairness in a datagram transport
> used among many peers in userspace is going to be timer resolution
> (especially on win32, but still problematic in *nix).
> 
> i'd much prefer to see a robust fractional bandwidth allocation
> specified for the datagram stack as a whole, than watch the poor
> behavior of another attempted TCP over UDP in userspace.
> 
> (i admit a general bias against TCP fairness in userspace, so a grain
> of salt with this $0.02..)

For the record, I'm pretty agnostic about which protocol to actually
use for reliable in-order delivery inside packets in a hypothetical
UDP Tor design.  I'd be surprised if the loss and latency
characteristics didn't turn out to have differences from those
typically encountered on the net at large, so it's quite possible that
something other than TCP will turn out to be the right choice.

This is going to be an exciting year for this topic, given the number
of people who seem to be interested in working on it: two currently
working on implementations that I'm aware of, plus the people on this
list taking an active interest.  It'll be neat when some of the
proof-of-concept implementations get to the stage where people can try
out different userspace protocols with them.  (It'll also be neat when
somebody actually writes up a detailed specification for their
protocol design in terms of a Proposal document. ;) )

yrs,
-- 
Nick