Thus spake Mike Perry (mikeperry@xxxxxxxxxx): > > > Given that a tor circuit is only as fast as its slowest node isn't it just as > > likely that for some reasonable percentage of the time, the user's circuit's > > will still be problematically slow (e.g. fast guard, slow relay, fast exit)? Oh, also, I missed pointing out another key assumption on balancing, since choice is weighted proportionally to bandwidth, increasing the bw capacity of these other nodes will cause users to choose them more often instead of the slower nodes, so it's not like the 10KB/sec nodes will just magically provide more bandwidth forever to support this rebalancing. The system will reach a new equilibrium, at which point more measurements will be needed to hunt down other, more subtle issues that are hiding behind the noise of these larger ones (such as liars, path biasers, potential uptime selection balancing issues, balancing directory vs tor traffic, etc). -- Mike Perry Mad Computer Scientist fscked.org evil labs
Attachment:
pgp84byVzj5ja.pgp
Description: PGP signature