Bittorrent manipulation by ISPs worldwide

Glasnost data from the beginning of 2009 to the end of Q1 2012


Find your Internet service provider, or any ISP of interest, in the “operator” selection bar. You can also view data by selecting a specific country code or time period (quarter). The table also shows the number and range of valid tests conducted. The column on the far right shows the percentage of tests indicating that the ISP was manipulating BitTorrent using DPI. Due to arbitrary variations in the measurements generated by the users’ computer or by noise/delays on the network,  Glasnost seems to have a 1 in 10 chance of generating a false positive (or false negative), so some ISPs who do not use DPI to throttle often show some positive percentage. In general, ISPs whose % of DPI is consistently below 10% over several consecutive quarters are unlikely to be doing anything; those whose results are consistently between 11% and 50% are likely to be using it in a limited way (e.g., confined to peak time periods or to specific networks or classes of users) while ISPs consistently over 50% are likely using it to manipulate nearly all BitTorrent traffic they encounter.

The number of valid tests is important because the more valid tests done, the more reliable the results in the last column. If there are less than 10 results per quarter, we do not show results for that ISP.

A series of interactive charts visualizing the results for the major ISPs in Canada, Germany, Great Britain, Poland, Taiwan, Hong Kong and Brazil is shown on another page.