Skip to main content

Why Is the Internet so Slow?!

  • Conference paper
  • First Online:
Passive and Active Measurement (PAM 2017)

Abstract

In principle, a network can transfer data at nearly the speed of light. Today’s Internet, however, is much slower: our measurements show that latencies are typically more than one, and often more than two orders of magnitude larger than the lower bound implied by the speed of light. Closing this gap would not only add value to today’s Internet applications, but might also open the door to exciting new applications. Thus, we propose a grand challenge for the networking research community: building a speed-of-light Internet. To help inform research towards this goal, we investigate, through large-scale measurements, the causes of latency inflation in the Internet across the network stack. Our analysis reveals an under-explored problem: the Internet’s infrastructural inefficiencies. We find that while protocol overheads, which have dominated the community’s attention, are indeed important, reducing latency inflation at the lowest layers will be critical for building a speed-of-light Internet. In fact, eliminating this infrastructural latency inflation, without any other changes in the protocol stack, could speed up small object fetches by more than a factor of three.

B. Chandrasekaran—This work was done when the author was a graduate student at Duke University.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 39.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 54.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    Data sets (gathered in 2016) and code are available at https://cgi.cs.duke.edu/~ilker/cspeed/pam2017-data/.

  2. 2.

    Explicit volunteer consent was obtained, listing precisely what tests would be run. We have a letter from the IRB stating that our tests did not require IRB approval.

  3. 3.

    We do not claim this is the percentage of Web sites supporting HTTPS.

References

  1. cURL. http://curl.haxx.se/

  2. ESnet. http://www.es.net/

  3. GÉANT. http://www.geant.net/

  4. Google Maps API. http://goo.gl/I4ypU

  5. Internet2. http://www.internet2.edu/

  6. RIPE Atlas. https://atlas.ripe.net

  7. Top 500 sites in each country or territory, Alexa. http://goo.gl/R8HuN6

  8. Workshop on reducing internet latency (2013). http://goo.gl/kQpBCt

  9. Akamai: State of the Internet, Q1 (2016). https://goo.gl/XQt324

  10. Brutlag, J.: Speed matters for Google Web search (2009). http://goo.gl/t7qGN8

  11. Dukkipati, N., Refice, T., Cheng, Y., Chu, J., Herbert, T., Agarwal, A., Jain, A., Sutin, N.: An argument for increasing TCP’s initial congestion window. In: SIGCOMM CCR (2010)

    Google Scholar 

  12. Durairajan, R., Barford, P., Sommers, J., Willinger, W.: Intertubes: a study of the US long-haul fiber-optic infrastructure. In: ACM SIGCOMM (2015)

    Google Scholar 

  13. Schurman, E., (Bing), Brutlag, J., (Google): Performance related changes and their user impact. http://goo.gl/hAUENq

  14. Gao, L., Wang, F.: The extent of AS path inflation by routing policies. In: GLOBECOM (2002)

    Google Scholar 

  15. Habib, M.A., Abrams, M.: Analysis of sources of latency in downloading web pages. In: WEBNET (2000)

    Google Scholar 

  16. Holterbach, T., Pelsser, C., Bush, R., Vanbever, L.: Quantifying interference between measurements on the RIPE Atlas platform (2015)

    Google Scholar 

  17. Grigorik, I., (Google): Latency: the new web performance bottleneck. http://goo.gl/djXp3

  18. Liddle, J.: Amazon Found Every 100ms of Latency Cost Them 1% in Sales. http://goo.gl/BUJgV

  19. Maynard-Koran, P.: Fixing the Internet for real time applications: Part II. http://goo.gl/46EiDC

  20. Mühlbauer, W., Uhlig, S., Feldmann, A., Maennel, O., Quoitin, B., Fu, B.: Impact of routing parameters on route diversity and path inflation. Comput. Netw. 54(14), 2506–2518 (2010)

    Article  MATH  Google Scholar 

  21. NEC: SEA-US: Global Consortium to Build Cable System Connecting Indonesia, the Philippines, and the United States. http://goo.gl/ZOV3qa

  22. Nordrum, A.: Fiber optics for the far North [News]. IEEE Spectr. 52(1), 11–13 (2015)

    Article  Google Scholar 

  23. Radhakrishnan, S., Cheng, Y., Chu, J., Jain, A., Raghavan, B.: TCP fast open. In: CoNEXT (2011)

    Google Scholar 

  24. Rexford, J., Wang, J., Xiao, Z., Zhang, Y.: BGP routing stability of popular destinations. In: ACM SIGCOMM Workshop on Internet Measurment (2002)

    Google Scholar 

  25. Singla, A., Chandrasekaran, B., Godfrey, P.B., Maggs, B.: The Internet at the speed of light. In: HotNets. ACM (2014)

    Google Scholar 

  26. Sundaresan, S., Magharei, N., Feamster, N., Teixeira, R.: Measuring and mitigating web performance bottlenecks in broadband access networks. In: IMC (2013)

    Google Scholar 

  27. Täht, D.: On reducing latencies below the perceptible. In: Workshop on Reducing Internet Latency (2013)

    Google Scholar 

  28. Vulimiri, A., Godfrey, P.B., Mittal, R., Sherry, J., Ratnasamy, S., Shenker, S.: Low latency via redundancy. In: CoNEXT (2013)

    Google Scholar 

  29. Wang, X.S., Balasubramanian, A., Krishnamurthy, A., Wetherall, D.: Demystify page load performance with WProf. In: NSDI (2013)

    Google Scholar 

  30. Wang, Z.: Speeding up mobile browsers without infrastructure support. Master’s thesis, Duke University (2012)

    Google Scholar 

  31. Zhou, W., Li, Q., Caesar, M., Godfrey, P.B.: ASAP: a low-latency transport layer. In: CoNEXT (2011)

    Google Scholar 

Download references

Acknowledgments

Dhruv Diddi helped process the ESnet data. Data on fiber mileages from GÉANT, the high-speed pan-European research and education network, was obtained through personal communication with Xavier Martins-Rivas, DANTE. DANTE is the project coordinator and operator of GÉANT.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ilker Nadi Bozkurt .

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2017 Springer International Publishing AG

About this paper

Cite this paper

Bozkurt, I.N. et al. (2017). Why Is the Internet so Slow?!. In: Kaafar, M., Uhlig, S., Amann, J. (eds) Passive and Active Measurement. PAM 2017. Lecture Notes in Computer Science(), vol 10176. Springer, Cham. https://doi.org/10.1007/978-3-319-54328-4_13

Download citation

  • DOI: https://doi.org/10.1007/978-3-319-54328-4_13

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-319-54327-7

  • Online ISBN: 978-3-319-54328-4

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics