Abstract

We present a series of experiments designed to test the effect of low bandwidth and asymmetrical bandwidth nodes on the performance of the Tor routing network. A simulation is designed to perform the tests, with a number of features that will make it useful for any further research into the performance of the Tor network. The experiments show that lower bandwidth does affect performance, and that there is a huge performance gap between having mostly and all slow nodes. We also find that asymmetrical connections perform no worse than symmetrical connections of equivalent upstream bandwidth. 1

Citations

2033	Congestion avoidance and control - Jacobson - 1988
523	Tor: The secondgeneration onion router - Dingledine, Mathewson, et al.
437	TCP vegas: New techniques for congestion detection and avoidance - Brakmo, O’Malley, et al. - 1994
212	Dynamics of IP traffic: A study of the role of variability and the impact of control - FELDMANN, GILBERT, et al. - 1999
177	Mixminion: Design of a type iii anonymous remailer protocol - Danezis, Dingledine, et al. - 2003
148	Mixing e-mail with BABEL - Gulcu, Tsudik - 1996
69	Nist net - a linux-based network emulation tool - Carson, Santay - 2003
66	The georgia tech network simulator - Riley - 2003
50	A behavioral model of Web traffic - Choi, Limb - 1999
36	Mapping peer behavior to packet-level details: A framework for packet-level simulation of peer-to-peer systems - He, Ammar, et al. - 2003
22	Network visualization with the VINT network animator Nam - Estrin, Handley, et al.
16	On the sensitivity of network simulation to topology - Anagnostakis, Greenwald, et al. - 2002
6	Challenges in deploying low-latency anonymity - Dingledine, Mathewson, et al. - 2005
2	ns Manual. A Collaboration between researchers at UC - Fall - 2006
1	A Collaboration between researchers at UC - unknown authors
1	Tutorial for the netowrk simulator ns - Greis