The Web, the Users, and the MOS: Influence of HTTP/2 on User Experience

  • Enrico Bocchi
  • Luca De Cicco
  • Marco Mellia
  • Dario Rossi
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10176)

Abstract

This work focuses on the evaluation of Web quality of experience as perceived by actual users and in particular on the impact of HTTP/1 vs HTTP/2. We adopt an experimental methodology that uses real web pages served through a realistic testbed where we control network, protocol, and application configuration. Users are asked to browse such pages and provide their subjective feedback, which we leverage to obtain the Mean Opinion Score (MOS), while the testbed records objective metrics.

The collected dataset comprises over 4,000 grades that we explore to tackle the question whether HTTP/2 improves users experience, to what extent, and in which conditions. Findings show that users report marginal differences, with 22%, 52%, 26% of HTTP/2 MOS being better, identical, or worse than HTTP/1, respectively. Even in scenarios that favor HTTP/2, results are not as sharp as expected. This is in contrast with objective metrics, which instead record a positive impact with HTTP/2 usage. This shows the complexity of understanding the web experience and the need to involve actual users in the quality assessment process.

Keywords

Web HTTP/2 Page Load Time MOS User experience QoE 

References

  1. 1.
    Allman, M., Paxson, V.: Issues and etiquette concerning use of shared measurement data. In: Proceedings of the 7th ACM SIGCOMM Conference on Internet Measurement, IMC 2007, pp. 135–140. ACM, New York (2007)Google Scholar
  2. 2.
    Bailey, M., Dittrich, D., Kenneally, E., Maughan, D.: The menlo report. IEEE Secur. Priv. 10(2), 71–75 (2012)CrossRefGoogle Scholar
  3. 3.
    Belshe, M., Peon, R., Thomson, M.: Hypertext transfer protocol version 2 (HTTP/2). In: IETF RFC7540 (2015)Google Scholar
  4. 4.
    Blackburn, B., Varvello, M., Schomp, K., Naylor, D., Finamore, A., Papagiannaki, K.: Is the Web HTTP/2 yet? In: TMA PhD School (2016)Google Scholar
  5. 5.
    Bocchi, E., De Cicco, L., Rossi, D.: Measuring the quality of experience of web users. In: Proceedings of ACM SIGCOMM Internet-QoE Workshop (2016)Google Scholar
  6. 6.
    Brutlag, J., Abrams, Z., Meenan, P.: Above the fold time: measuring web page performance visually. http://conferences.oreilly.com/velocity/velocity-mar2011/public/schedule/detail/18692. Accessed 15 Sept 2016
  7. 7.
    Butkiewicz, M., Wang, D., Wu, Z., Madhyastha, H.V., Sekar, V.: Klotski: reprioritizing web content to improve user experience on mobile devices. In: Proceedings of USENIX NSDI, pp. 439–453 (2015)Google Scholar
  8. 8.
    Erman, J., Gopalakrishnan, V., Jana, R., Ramakrishnan, K.K.: Towards a SPDY’Ier Mobile Web? In: Proceedings of ACM CoNEXT, pp. 303–314 (2013)Google Scholar
  9. 9.
  10. 10.
    Google Inc. QUIC. https://www.chromium.org/quic. Accessed 19 Sept 2016
  11. 11.
    Google Inc. SPDY. https://www.chromium.org/spdy/spdy-whitepaper. Accessed 19 Sept 2016
  12. 12.
    Grigorik, I.: HTTP/2 is here, let’s optimize! http://bit.ly/http2-opt. Accessed 10 Oct 2016
  13. 13.
    International Telecommunication Union. Subjective testing methodology for web browsing. ITU-T Recommendation P.1501 (2014)Google Scholar
  14. 14.
    Irish, P.: Delivering the goods in under 1000 ms. http://bit.ly/1toUUA7 Accessed 10 Oct 2016
  15. 15.
    Varvello, M., Schomp, K., Naylor, D., Blackburn, J., Finamore, A., Papagiannaki, K.: Is the web HTTP/2 yet? In: Karagiannis, T., Dimitropoulos, X. (eds.) PAM 2016. LNCS, vol. 9631, pp. 218–232. Springer, Heidelberg (2016). doi:10.1007/978-3-319-30505-9_17 CrossRefGoogle Scholar
  16. 16.
    Miller, R.B.: Response time in man-computer conversational transactions. In: Proceedings of AFIPS Fall Joint Computer Conference, pp. 267–277 (1968)Google Scholar
  17. 17.
    Netravali, R., Sivaraman, A., Das, S., Goyal, A., Winstein, K., Mickens, J., Balakrishnan, H.: Mahimahi: accurate record-and-replay for HTTP. In: Proceedings of USENIX ATC, pp. 417–429 (2015)Google Scholar
  18. 18.
    Nielsen, J.: Response times: the 3 important limits.https://www.nngroup.com/articles/response-times-3-important-limits/ (1993). Accessed 19 Sept 2016
  19. 19.
    Popa, L., Ghodsi, A., Stoica, I.: HTTP as the narrow waist of the future internet. In: 9th ACM SIGCOMM Workshop on Hot Topics in Networks (2010)Google Scholar
  20. 20.
    Reichl, P., Egger, S., Schatz, R., D’Alconzo, A.: The logarithmic nature of QoE and the role of the Weber-Fechner law in QoE assessment. In: IEEE ICC (2010)Google Scholar
  21. 21.
    Wang, X.S., Balasubramanian, A., Krishnamurthy, A., Wetherall, D.: How speedy is SPDY? In: Proceedings of USENIX NSDI, pp. 387–399 (2014)Google Scholar
  22. 22.
    Wang, X.S., Krishnamurthy, A., Wetherall, D.: Speeding up web page loads with Shandian. In: Proceedings of USENIX NSDI, pp. 109–122 (2016)Google Scholar
  23. 23.
    Zarifis, K., Holland, M., Jain, M., Katz-Bassett, E., Govindan, R.: Modeling HTTP/2 speed from HTTP/1 traces. In: Karagiannis, T., Dimitropoulos, X. (eds.) PAM 2016. LNCS, vol. 9631, pp. 233–247. Springer, Heidelberg (2016). doi:10.1007/978-3-319-30505-9_18 CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Enrico Bocchi
    • 1
  • Luca De Cicco
    • 2
  • Marco Mellia
    • 3
  • Dario Rossi
    • 4
  1. 1.Télécom ParisTechParisFrance
  2. 2.Politecnico di BariBariItaly
  3. 3.Politecnico di TorinoTorinoItaly
  4. 4.Ecole Nationale Supérieure des TélécommunicationsParisFrance

Personalised recommendations