On 10/04/2014 05:57 PM, Virgil Griffith wrote: > I hereby submit the following PDF as a Tor Tech Report. > > http://dl.dropbox.com/u/3308162/tor%20growth.pdf > > I'll be providing the various raw data files to github/torproject shortly. > > Qualitative conclusions are the same. The methodology is more > rigorous than it was before, some quantitative differences. Most > notably: > * Lots of stuff in Tor increases exponentially. > * Absolute Torperf is increasingly nicely. > * Normalized Torperf (Torperf relative to non-tor speeds) increases glacially. > * There's statistical evidence that an injection of advertised > bandwidth would reliably improve Torperf performance. > > For non-policy makers, I think the main contribution I am offering is > a list of the various Tor growth rates (page 4, table 2)
I've noticed something relevant to current discussion about Tor use. | Figure 8: Network Utilization Ratio (NUR) falls into three distinct | stages. Within each stage the fitted line is essentially flat. What | happened on 2013-10-08 and 2014-06-06!? The only thing we see is | that on 2014-06-05 (one day prior) the EFF began their Tor Challenge. Silk Road was taken down by the FBI on 2014-10-02.[0] The site still loaded, however, showing a "This Hidden Site Has Been Seized" message. And so it probably took a few days for most users to get the news. Also, contemplating Figure 8, there may be a discontinuity at about July 2013, when they took down Freedom Hosting.[1] I'm not aware of anything major in May-June of this year. [0] https://lists.torproject.org/pipermail/tor-talk/2013-October/030268.html [1] http://www.wired.com/2013/09/freedom-hosting-fbi/ -- tor-talk mailing list - [email protected] To unsubscribe or change other settings go to https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk
