Releases data collected from more than 8000 users to show that Comcast isn’t the only interfering with BitTorrent traffic.
About a month ago I mentioned how Azureus was asking users of the BitTorrent client to install a simple software plug-in in order to help it “…gather information about potential interference with your Internet traffic.” Specifically, Azureus wanted to collect data showing just how widespread the practice of BitTorrent throttling might be amongst the world’s ISPs.
Well, now the results are in and the findings point to a more widespread practice of BitTorrent throttling than previously contemplated. To date, some 8000 users around the world have opted to install the plug-in, providing over 1,000,000 hours of data for it to analyze.
The plug in constantly monitored the rate of network interruptions occurring from RST (“reset”) packets by measuring the total number of attempted network connections and the total number of network connections that were interrupted by a reset message. Comparing these two values, it calculated the ratio of network connections interrupted by reset messages.
So Who Are the Worst Offenders?
1. Comcast USA 23.72%
2. Cogeco Canada 19.13%
3. Emirates Internet UAE 17.86%
4. Cablevision USA 17.58%
5. Brasil Telecom Santa Catarina, Brazil 17.43%
6. TM Net Malaysia 16.80%
7. BellSouth USA 15.88%
8. Tedata Egypt 15.33%
9. Tiscali UK 14.89%
10. AOL USA 14.88%
11. Banda Ancha SA 14.44%
12. AtHome Benelux BV 14.12%
13. AT&T World Service 13.97%
14. Charter Communications 13.20%
15. AvatarBrodband Limited 13.19%
16. PIPEX Communications 13.16%
17. Rogers Cable Communications Inc. 13.07%
18. Uninet S.A. de C.V. 12.6%
19. TekSavvy Solutions, Inc 12.48%
20. Qwest 12.3%
1. Telecom Italia France 2.53%
2. Orange Nederland 2.57%
3. WiLine Networks Inc.2.78%
4. Telefonica Deutschland Autonomous System 3.60%
5. Freenet Cityline GmbH 4.21%
Azureus notes that while it cannot definitively conclude that any particular ISP is engaging in artificial or false RST packet behavior(BitTorrent throttling) it is possible to compare the experience of different BitTorrent users in different networks. It is because of this comparison that it appears that the rate of interrupted network communications is disproportionately high in some ISPs versus others.
We believe that there is sufficient data to suggest that network management practices that “throttle” internet traffic are widespread. At a minimum, more investigation is required to determine whether these resets are happening in the ordinary course of business or whether they represent the kind of throttling practices which target specific applications and/or protocols, harming the consumer experience and stifling innovation.
For more information to take a look at and compare to these plug in results, there’s a great Wiki page that’s been around for a while that lists ISPs that are “…known to cause trouble for BitTorrent clients or other P2P clients and the reason why. “