On Tue, Jan 17, 2017 at 05:11:12PM +0100, Karsten Loesing wrote:
Firstly, while we don't have (public) datasets showing this, I am fairly certain that the public meek bridges have their bandwidth capped (to limit costs) and raising that cap would directly improve the speed of meek users. (Implying that the bottleneck the majority of the time is the meek transport.) Right?
That's also my intuition, but I don't have the data you're looking for. David would know better.
You can probably look for some patterns in past bandwidth history and compare them to the history of rate limit changes at https://trac.torproject.org/projects/tor/wiki/doc/meek#Users
Active bridges: https://atlas.torproject.org/#details/AA033EEB61601B2B7312D89B62AAA23DC3ED8A... meek-azure https://atlas.torproject.org/#details/F4AD82B2032EDEF6C02C5A529C42CFAFE51656... meek-amazon (after November 2015) Defunct bridge fingerprints: https://atlas.torproject.org/#details/3FD131B74D9A96190B1EE5D31E91757FADA1A4... meek-amazon (before November 2015; #17473) https://atlas.torproject.org/#details/88F745840F47CE0C6A4FE61D827950B06F9E45... meek-google (before May 2016)
For example, take * 2015-10-02 Rate-limited the meek-azure bridge to 1.1 MB/s. (Azure grant expired.) and you can see a matching drop in bandwidth from 2.5 MB/s to 1.1 MB/s on the same date. Also, * 2016-01-14 Increased meek-azure rate limit to 3 MB/s. * 2016-01-16 Increased meek-amazon rate limit to 3 MB/s. is really visible in the bandwidth graphs.