[tor-relays] does the Tor network scale well ?

nusenu nusenu at openmailbox.org
Mon Feb 22 22:43:02 UTC 2016



Toralf Förster:
> Given the numbers of https://torstatus.blutmagie.de/index.php there
> are about 40 Tor exits which covers 1/3 of the whole bandwidth of all
> exit relays. I do wonder if under-weighting the big relays in favour
> of many smaller relays would make it harder for an adversary to
> de-anonymize the traffic ?


A more diverse tor network (more relays, in more locations, and a more
balanced cw-fraction across relays) would probably make it harder for an
adversary which can only observe a few spots on the internet to watch
most of the tor traffic. Manipulating CW values makes it certainly
possible to distribute traffic more evenly but that comes with a
performance impact. How bad is it really and how much are you willing to
"pay" (in latency+bw) to make it how much better?

In the end its usually performance vs. anonymity properties. Bad
performance results in fewer users, and smaller anonymity sets imply
weaker anonymity properties.

Unfortunately there are no incentives for geo or AS diversity or even
limits like "any given hoster/AS/country/operator.. is not allowed to
host more than ___% of the tor network".

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: OpenPGP digital signature
URL: <http://lists.torproject.org/pipermail/tor-relays/attachments/20160222/6ad7216e/attachment.sig>


More information about the tor-relays mailing list