The path less travelled

Overcoming Tor's bottlenecks with traffic splitting

Mashael Alsabah, Kevin Bauer, Tariq Elahi, Ian Goldberg

Research output: Chapter in Book/Report/Conference proceedingConference contribution

17 Citations (Scopus)

Abstract

Tor is the most popular low-latency anonymity network for enhancing ordinary users' online privacy and resisting censorship. While it has grown in popularity, Tor has a variety of performance problems that result in poor quality of service, a strong disincentive to use the system, and weaker anonymity properties for all users. We observe that one reason why Tor is slow is due to low-bandwidth volunteer-operated routers. When clients use a low-bandwidth router, their throughput is limited by the capacity of the slowest node. With the introduction of bridges - unadvertised Tor routers that provide Tor access to users within censored regimes like China - low-bandwidth Tor routers are becoming more common and essential to Tor's ability to resist censorship. In this paper, we present Conflux, a dynamic traffic-splitting approach that assigns traffic to an overlay path based on its measured latency. Because it enhances the load-balancing properties of the network, Conflux considerably increases performance for clients using low-bandwidth bridges. Moreover, Conflux significantly improves the experience of users who watch streaming videos online. Through live measurements and a whole-network evaluation conducted on a scalable network emulator, we show that our approach offers an improvement of approximately 30% in expected download time for web browsers who use Tor bridges and for streaming application users. We also show that Conflux introduces only slight tradeoffs between users' anonymity and performance.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages143-163
Number of pages21
Volume7981 LNCS
DOIs
Publication statusPublished - 2013
Externally publishedYes
Event13th International Symposium on Privacy Enhancing Technologies, PETS 2013 - Bloomington, IN
Duration: 10 Jul 201312 Jul 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7981 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other13th International Symposium on Privacy Enhancing Technologies, PETS 2013
CityBloomington, IN
Period10/7/1312/7/13

Fingerprint

Routers
Traffic
Router
Bandwidth
Path
Anonymity
Web browsers
Latency
Video streaming
Resource allocation
Traffic Dynamics
Quality of service
Video Streaming
Throughput
Overlay
Streaming
Load Balancing
Resist
Quality of Service
Privacy

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Alsabah, M., Bauer, K., Elahi, T., & Goldberg, I. (2013). The path less travelled: Overcoming Tor's bottlenecks with traffic splitting. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7981 LNCS, pp. 143-163). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7981 LNCS). https://doi.org/10.1007/978-3-642-39077-7_8

The path less travelled : Overcoming Tor's bottlenecks with traffic splitting. / Alsabah, Mashael; Bauer, Kevin; Elahi, Tariq; Goldberg, Ian.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7981 LNCS 2013. p. 143-163 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7981 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Alsabah, M, Bauer, K, Elahi, T & Goldberg, I 2013, The path less travelled: Overcoming Tor's bottlenecks with traffic splitting. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 7981 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7981 LNCS, pp. 143-163, 13th International Symposium on Privacy Enhancing Technologies, PETS 2013, Bloomington, IN, 10/7/13. https://doi.org/10.1007/978-3-642-39077-7_8
Alsabah M, Bauer K, Elahi T, Goldberg I. The path less travelled: Overcoming Tor's bottlenecks with traffic splitting. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7981 LNCS. 2013. p. 143-163. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-39077-7_8
Alsabah, Mashael ; Bauer, Kevin ; Elahi, Tariq ; Goldberg, Ian. / The path less travelled : Overcoming Tor's bottlenecks with traffic splitting. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7981 LNCS 2013. pp. 143-163 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{64ecd521113f41e69a3bbf8f0f9479d0,
title = "The path less travelled: Overcoming Tor's bottlenecks with traffic splitting",
abstract = "Tor is the most popular low-latency anonymity network for enhancing ordinary users' online privacy and resisting censorship. While it has grown in popularity, Tor has a variety of performance problems that result in poor quality of service, a strong disincentive to use the system, and weaker anonymity properties for all users. We observe that one reason why Tor is slow is due to low-bandwidth volunteer-operated routers. When clients use a low-bandwidth router, their throughput is limited by the capacity of the slowest node. With the introduction of bridges - unadvertised Tor routers that provide Tor access to users within censored regimes like China - low-bandwidth Tor routers are becoming more common and essential to Tor's ability to resist censorship. In this paper, we present Conflux, a dynamic traffic-splitting approach that assigns traffic to an overlay path based on its measured latency. Because it enhances the load-balancing properties of the network, Conflux considerably increases performance for clients using low-bandwidth bridges. Moreover, Conflux significantly improves the experience of users who watch streaming videos online. Through live measurements and a whole-network evaluation conducted on a scalable network emulator, we show that our approach offers an improvement of approximately 30{\%} in expected download time for web browsers who use Tor bridges and for streaming application users. We also show that Conflux introduces only slight tradeoffs between users' anonymity and performance.",
author = "Mashael Alsabah and Kevin Bauer and Tariq Elahi and Ian Goldberg",
year = "2013",
doi = "10.1007/978-3-642-39077-7_8",
language = "English",
isbn = "9783642390760",
volume = "7981 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "143--163",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - The path less travelled

T2 - Overcoming Tor's bottlenecks with traffic splitting

AU - Alsabah, Mashael

AU - Bauer, Kevin

AU - Elahi, Tariq

AU - Goldberg, Ian

PY - 2013

Y1 - 2013

N2 - Tor is the most popular low-latency anonymity network for enhancing ordinary users' online privacy and resisting censorship. While it has grown in popularity, Tor has a variety of performance problems that result in poor quality of service, a strong disincentive to use the system, and weaker anonymity properties for all users. We observe that one reason why Tor is slow is due to low-bandwidth volunteer-operated routers. When clients use a low-bandwidth router, their throughput is limited by the capacity of the slowest node. With the introduction of bridges - unadvertised Tor routers that provide Tor access to users within censored regimes like China - low-bandwidth Tor routers are becoming more common and essential to Tor's ability to resist censorship. In this paper, we present Conflux, a dynamic traffic-splitting approach that assigns traffic to an overlay path based on its measured latency. Because it enhances the load-balancing properties of the network, Conflux considerably increases performance for clients using low-bandwidth bridges. Moreover, Conflux significantly improves the experience of users who watch streaming videos online. Through live measurements and a whole-network evaluation conducted on a scalable network emulator, we show that our approach offers an improvement of approximately 30% in expected download time for web browsers who use Tor bridges and for streaming application users. We also show that Conflux introduces only slight tradeoffs between users' anonymity and performance.

AB - Tor is the most popular low-latency anonymity network for enhancing ordinary users' online privacy and resisting censorship. While it has grown in popularity, Tor has a variety of performance problems that result in poor quality of service, a strong disincentive to use the system, and weaker anonymity properties for all users. We observe that one reason why Tor is slow is due to low-bandwidth volunteer-operated routers. When clients use a low-bandwidth router, their throughput is limited by the capacity of the slowest node. With the introduction of bridges - unadvertised Tor routers that provide Tor access to users within censored regimes like China - low-bandwidth Tor routers are becoming more common and essential to Tor's ability to resist censorship. In this paper, we present Conflux, a dynamic traffic-splitting approach that assigns traffic to an overlay path based on its measured latency. Because it enhances the load-balancing properties of the network, Conflux considerably increases performance for clients using low-bandwidth bridges. Moreover, Conflux significantly improves the experience of users who watch streaming videos online. Through live measurements and a whole-network evaluation conducted on a scalable network emulator, we show that our approach offers an improvement of approximately 30% in expected download time for web browsers who use Tor bridges and for streaming application users. We also show that Conflux introduces only slight tradeoffs between users' anonymity and performance.

UR - http://www.scopus.com/inward/record.url?scp=84884955109&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884955109&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-39077-7_8

DO - 10.1007/978-3-642-39077-7_8

M3 - Conference contribution

SN - 9783642390760

VL - 7981 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 143

EP - 163

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -