Shouldn't there be a way to set up an isolated mini-Tor network with just a few high-bandwidth nodes for speed testing? Sort of like a lab test where no traffic reaches the real Tor network (sort of a sandbox)? Or is mimicking the Directories and such just too much work? best, Joe On Sunday, February 27, 2011, raidz <admin@xxxxxxxxx> wrote: > I have a feeling the issue lies in the code, I mean Tor isn't even > multi-threaded, thats pathetic. I have a relay that maxes out at 100mbits on > a gige line, it won't go any higher. It would be great if we could work > together with the developers, possibly giving them access to some of our > relays and exits to see if they can find where the bottleneck lies. With > people getting more used to faster speeds it will turn some people off from > using the network. I know speed isn't the first priority, but it should be > one of the higher priorities. > > Moritz, > > I am not sure who would have better luck with the devs, I can try and reach > out to them, although I know you probably have a better relationship with > them. I would be more than happy to help them troubleshoot any issues, I just > hope they have the time and are willing because as it is now, Tor is not > scalable. > > > On Sun, 27 Feb 2011 21:54:28 +0100, Moritz Bartl wrote: > > Hi, > > Am 27.02.2011 21:43, schrieb Mark Smith: > > Could it simply be that the relays are offering more bandwidth than users > need? > > > That question comes up from time to time. I don't know the answer, and I > surely would like to know more. > > I cannot believe that there ever will be a point in time where all our > hundredthousands of users are unable to saturate the Tor network . When > you use Tor, you can clearly see it could use more speed. And that's not > only the latency of 3 hops. > > > Still, the exits may not be a limiting factor anymore. > > > The last time I ran the script to generate the tormap for Google Maps, > only around 80 exits were faster than 1 mb/s. That includes a number of > "more logical Tor nodes on one physical line because Tor does not > scale", and many exits who only allow a very limited set of ports to exit. > One should be aware that being an exit relay is no different from being > a middle relay, only *additionally* allowing traffic to exit. So even if > exits were not the bottleneck, they should still be used as middle and > entry nodes. > > > > -- Joseph Lorenzo Hall ACCURATE Postdoctoral Research Associate UC Berkeley School of Information Princeton Center for Information Technology Policy http://josephhall.org/