Hi Roger, I looked at traffic from the latest Orbot (2.3.10-alpha) vs an older release (2.2.14-alpha) and found that idle usage was reduced. The latest Orbot used ~65 bytes/s while the older version used ~91 bytes/sec. An idle data reduction of around 30%. Does that sound reasonable?
That blog post now has incorrect statistics so I will update it to reflect the lower usage. Thanks, Patrick On Sat, Jun 23, 2012 at 7:04 PM, Roger Dingledine <a...@mit.edu> wrote: > On Wed, Jun 20, 2012 at 05:13:16PM -0400, Patrick B wrote: > > I made a blog post < > https://guardianproject.info/2012/06/20/orbot-data-tax/> on > > the Guardian site about the data cost incurred by Orbot usage. It > generally > > seems quite manageable for most use cases. Running in the backround 24/7 > > for 30 days does incur ~200MB of idle usage which would be significant > for > > mobile users with smaller data plans. Check out the post for more > details. > > I'm also curious if anyone has taken similar measurements with Tor to > > compare the results. > > Hi Patrick, > > You should compare Tor 0.2.2.x with Tor 0.2.3.x in terms of idle bandwidth > load. Tor 0.2.3.x uses microdescriptors rather than descriptors for > directory information, so it should reduce the daily amount of stuff > your Tor has to fetch to keep informed about what relays are available. > > > https://gitweb.torproject.org/torspec.git/blob/HEAD:/proposals/158-microdescriptors.txt > > Also, if your Tor stays idle for long periods of time (like, days), > you might find https://trac.torproject.org/projects/tor/ticket/2149 > interesting (not yet implemented). > > --Roger > > _______________________________________________ > tor-talk mailing list > tor-talk@lists.torproject.org > https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk > _______________________________________________ tor-talk mailing list tor-talk@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk
