Fury Route

Leveraging CDNs to Remotely Measure Network Distance

Marcel Flores*, Alexander Wenzel, Kevin Chen, Aleksandar Kuzmanovic

*Corresponding author for this work

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

1 Citation (Scopus)

Abstract

Estimating network distance between arbitrary Internet endpoints is an essential primitive in applications ranging from performance optimization to network debugging and auditing. Enabling such a primitive without deploying new infrastructure was demonstrated via DNS. However, the proliferation of DNS hosting has made DNS-based measurement techniques far less dependable. In this paper, we show that the heterogeneous infrastructure of different CDNs, combined with the proliferation of the EDNS0 client-subnet extension (ECS), enables novel infrastructureless measurement. We design Fury Route, a system that estimates network distance by utilizing ECS to construct a virtual path between endpoints via intermediate CDN replicas. Fury Route requires no additional infrastructure to be deployed. The measured endpoints do not need to participate by sending or responding to probes. Fury Route further generates no load on endpoints. It only queries DNS, whose infrastructure is designed for large loads. We extensively evaluate Fury Route and demonstrate that (i) the key to Fury Route’s ability to construct virtual paths lies in the heterogeneity of the underlying CDNs, (ii) Fury Route is effective in revealing relative network distance, needed in many real-world scenarios, (iii) caching can dramatically reduce Fury Route’s DNS overhead, making it a useful system in practice.

Original languageEnglish (US)
Title of host publicationPassive and Active Measurement - 19th International Conference, PAM 2018, Proceedings
EditorsAnja Feldmann, Georgios Smaragdakis, Robert Beverly
PublisherSpringer Verlag
Pages87-99
Number of pages13
ISBN (Print)9783319764801
DOIs
StatePublished - Jan 1 2018
Event19th International Conference on Passive and Active Measurement, PAM 2018 - Berlin, Germany
Duration: Mar 26 2018Mar 27 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10771 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other19th International Conference on Passive and Active Measurement, PAM 2018
CountryGermany
CityBerlin
Period3/26/183/27/18

Fingerprint

Infrastructure
Proliferation
Internet
Auditing
Path
Performance Optimization
Measurement Techniques
Caching
Debugging
Replica
Probe
Direct numerical Simulation
Query
Scenarios
Evaluate
Arbitrary
Estimate
Demonstrate

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Flores, M., Wenzel, A., Chen, K., & Kuzmanovic, A. (2018). Fury Route: Leveraging CDNs to Remotely Measure Network Distance. In A. Feldmann, G. Smaragdakis, & R. Beverly (Eds.), Passive and Active Measurement - 19th International Conference, PAM 2018, Proceedings (pp. 87-99). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10771 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-76481-8_7
Flores, Marcel ; Wenzel, Alexander ; Chen, Kevin ; Kuzmanovic, Aleksandar. / Fury Route : Leveraging CDNs to Remotely Measure Network Distance. Passive and Active Measurement - 19th International Conference, PAM 2018, Proceedings. editor / Anja Feldmann ; Georgios Smaragdakis ; Robert Beverly. Springer Verlag, 2018. pp. 87-99 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{e8ab73bcb3124ffe994929d90fd9070a,
title = "Fury Route: Leveraging CDNs to Remotely Measure Network Distance",
abstract = "Estimating network distance between arbitrary Internet endpoints is an essential primitive in applications ranging from performance optimization to network debugging and auditing. Enabling such a primitive without deploying new infrastructure was demonstrated via DNS. However, the proliferation of DNS hosting has made DNS-based measurement techniques far less dependable. In this paper, we show that the heterogeneous infrastructure of different CDNs, combined with the proliferation of the EDNS0 client-subnet extension (ECS), enables novel infrastructureless measurement. We design Fury Route, a system that estimates network distance by utilizing ECS to construct a virtual path between endpoints via intermediate CDN replicas. Fury Route requires no additional infrastructure to be deployed. The measured endpoints do not need to participate by sending or responding to probes. Fury Route further generates no load on endpoints. It only queries DNS, whose infrastructure is designed for large loads. We extensively evaluate Fury Route and demonstrate that (i) the key to Fury Route’s ability to construct virtual paths lies in the heterogeneity of the underlying CDNs, (ii) Fury Route is effective in revealing relative network distance, needed in many real-world scenarios, (iii) caching can dramatically reduce Fury Route’s DNS overhead, making it a useful system in practice.",
author = "Marcel Flores and Alexander Wenzel and Kevin Chen and Aleksandar Kuzmanovic",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-319-76481-8_7",
language = "English (US)",
isbn = "9783319764801",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "87--99",
editor = "Anja Feldmann and Georgios Smaragdakis and Robert Beverly",
booktitle = "Passive and Active Measurement - 19th International Conference, PAM 2018, Proceedings",
address = "Germany",

}

Flores, M, Wenzel, A, Chen, K & Kuzmanovic, A 2018, Fury Route: Leveraging CDNs to Remotely Measure Network Distance. in A Feldmann, G Smaragdakis & R Beverly (eds), Passive and Active Measurement - 19th International Conference, PAM 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10771 LNCS, Springer Verlag, pp. 87-99, 19th International Conference on Passive and Active Measurement, PAM 2018, Berlin, Germany, 3/26/18. https://doi.org/10.1007/978-3-319-76481-8_7

Fury Route : Leveraging CDNs to Remotely Measure Network Distance. / Flores, Marcel; Wenzel, Alexander; Chen, Kevin; Kuzmanovic, Aleksandar.

Passive and Active Measurement - 19th International Conference, PAM 2018, Proceedings. ed. / Anja Feldmann; Georgios Smaragdakis; Robert Beverly. Springer Verlag, 2018. p. 87-99 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10771 LNCS).

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

TY - GEN

T1 - Fury Route

T2 - Leveraging CDNs to Remotely Measure Network Distance

AU - Flores, Marcel

AU - Wenzel, Alexander

AU - Chen, Kevin

AU - Kuzmanovic, Aleksandar

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Estimating network distance between arbitrary Internet endpoints is an essential primitive in applications ranging from performance optimization to network debugging and auditing. Enabling such a primitive without deploying new infrastructure was demonstrated via DNS. However, the proliferation of DNS hosting has made DNS-based measurement techniques far less dependable. In this paper, we show that the heterogeneous infrastructure of different CDNs, combined with the proliferation of the EDNS0 client-subnet extension (ECS), enables novel infrastructureless measurement. We design Fury Route, a system that estimates network distance by utilizing ECS to construct a virtual path between endpoints via intermediate CDN replicas. Fury Route requires no additional infrastructure to be deployed. The measured endpoints do not need to participate by sending or responding to probes. Fury Route further generates no load on endpoints. It only queries DNS, whose infrastructure is designed for large loads. We extensively evaluate Fury Route and demonstrate that (i) the key to Fury Route’s ability to construct virtual paths lies in the heterogeneity of the underlying CDNs, (ii) Fury Route is effective in revealing relative network distance, needed in many real-world scenarios, (iii) caching can dramatically reduce Fury Route’s DNS overhead, making it a useful system in practice.

AB - Estimating network distance between arbitrary Internet endpoints is an essential primitive in applications ranging from performance optimization to network debugging and auditing. Enabling such a primitive without deploying new infrastructure was demonstrated via DNS. However, the proliferation of DNS hosting has made DNS-based measurement techniques far less dependable. In this paper, we show that the heterogeneous infrastructure of different CDNs, combined with the proliferation of the EDNS0 client-subnet extension (ECS), enables novel infrastructureless measurement. We design Fury Route, a system that estimates network distance by utilizing ECS to construct a virtual path between endpoints via intermediate CDN replicas. Fury Route requires no additional infrastructure to be deployed. The measured endpoints do not need to participate by sending or responding to probes. Fury Route further generates no load on endpoints. It only queries DNS, whose infrastructure is designed for large loads. We extensively evaluate Fury Route and demonstrate that (i) the key to Fury Route’s ability to construct virtual paths lies in the heterogeneity of the underlying CDNs, (ii) Fury Route is effective in revealing relative network distance, needed in many real-world scenarios, (iii) caching can dramatically reduce Fury Route’s DNS overhead, making it a useful system in practice.

UR - http://www.scopus.com/inward/record.url?scp=85043588353&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043588353&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-76481-8_7

DO - 10.1007/978-3-319-76481-8_7

M3 - Conference contribution

SN - 9783319764801

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

SP - 87

EP - 99

BT - Passive and Active Measurement - 19th International Conference, PAM 2018, Proceedings

A2 - Feldmann, Anja

A2 - Smaragdakis, Georgios

A2 - Beverly, Robert

PB - Springer Verlag

ER -

Flores M, Wenzel A, Chen K, Kuzmanovic A. Fury Route: Leveraging CDNs to Remotely Measure Network Distance. In Feldmann A, Smaragdakis G, Beverly R, editors, Passive and Active Measurement - 19th International Conference, PAM 2018, Proceedings. Springer Verlag. 2018. p. 87-99. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-76481-8_7