Modeling HTTP/2 Speed from HTTP/1 Traces

  • Kyriakos ZarifisEmail author
  • Mark Holland
  • Manish Jain
  • Ethan Katz-Bassett
  • Ramesh Govindan
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9631)


With the standardization of HTTP/2, content providerswant to understand the benefits and pitfalls of transitioning to the new standard. Using a large dataset of HTTP/1.1 resource timing data from production traffic on Akamai’s CDN, and a model of HTTP/2 behavior, we obtain the distribution of performance differences between the protocol versions for nearly 280,000 downloads. We find that HTTP/2 provides significant performanceimprovements in the tail, and, for websites for which HTTP/2 does not improve median performance, we explore how optimizations like prioritization and push can improve performance, and how these improvements relate to page structure.


Critical Path Content Provider Congestion Window Parallel Connection Page View 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.



We thank our shepherd, Srikanth Sundaresan, and the reviewers for their helpful comments. Kyriakos Zarifis performed this work while employed temporarily at Akamai. This work was funded in part by the National Science Foundation (NSF) under grant number CNS-1413978.


  1. 1.
    HTTP Pipelining Not So Fast (Nor Slow!).
  2. 2.
    Resource Timing Specification.
  3. 3.
  4. 4.
  5. 5.
    Cherif, W., Fablet, Y., Nassor, E., Taquet, J., Fujimori, Y.: Dash fast start using HTTP/2. In: NOSSDAV (2015)Google Scholar
  6. 6.
    El-Khatib, Y., Tyson, G., Welzl, M.: Can SPDY really make the web faster? In: IFIP Networking Conference (2014)Google Scholar
  7. 7.
    Erman, J., Gopalakrishnan, V., Jana, R., Ramakrishnan, K.K.: Towards a SPDY’ier mobile web? In: CoNEXT (2013)Google Scholar
  8. 8.
    Flach, T., Dukkipati, N., Terzis, A., Raghavan, B., Cardwell, N., Cheng, Y., Jain, A., Hao, S., Katz-Bassett, E., Govindan, R.: Reducing web latency: the virtue of gentle aggression. In: SIGCOMM (2013)Google Scholar
  9. 9.
    Ha, S., Rhee, I., Xu, L.: CUBIC: a new tcp-friendly high-speed TCP variant. Operating Syst. Rev. 42, 64–74 (2008)CrossRefGoogle Scholar
  10. 10.
    Meenan, P.: How fast is your web site? Commun. ACM 56, 49–55 (2013)CrossRefGoogle Scholar
  11. 11.
    Padhye, J., Nielsen, H.F.: A comparison of SPDY and HTTP performance. Technical report, July 2012Google Scholar
  12. 12.
    Varvello, M., Schomp, K., Naylor, D., Blackburn, J., Finamore, A., Papagiannaki, K.: To HTTP/2, or not to HTTP/2, that is the question. In: PAM (2016)Google Scholar
  13. 13.
    Wang, X.S., Balasubramanian, A., Krishnamurthy, A., Wetherall, D.: Demystifying page load performance with wprof. In: NSDI (2013)Google Scholar
  14. 14.
    Wang, X.S., Balasubramanian, A., Krishnamurthy, A., Wetherall, D.: How speedy is SPDY? In: NSDI (2014)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Kyriakos Zarifis
    • 1
    Email author
  • Mark Holland
    • 2
  • Manish Jain
    • 2
  • Ethan Katz-Bassett
    • 1
  • Ramesh Govindan
    • 1
  1. 1.University of Southern CaliforniaLos AngelesUSA
  2. 2.Akamai TechnologiesCambridgeUSA

Personalised recommendations