Understanding the causal impact of the video delivery throughput on user engagement

  • Qingxia LiEmail author


In this paper, we first examine the causal relationship between the perceived video download speed and user engagement, while the speed has a very limited impact on views of short-length videos, a lower speed could significantly impair the viewing completion rate of medium and long-length videos. In addition, we observe that a view with an experienced speed close to half of the video bitrate lasts 10% of the video less in comparison to a similar view with a higher relative speed. At last, We also pointed out that crossing AS (Autonomous System) borders does not necessarily imply a higher likelihood of being a problem session, as only 22% of the AS pairs show a statistically significant impact on video download speed.


Multimedia Video delivery Throughput User engagement 



  1. 1.
    Alvim SDSD, Borges VA, Ribeiro-Neto B, Vale ACS (2003) Performance analysis and optimization of a distributed video on demand service. IEEE Int Sym Perform Anal Syst Softw ISPASS 2003:156–165Google Scholar
  2. 2.
    Balachandran A, Sekar V, Akella A, Seshan S, Stoica I, Zhang H (2013) Developing a predictive model of quality of experience for internet video. Comput Commun Rev 43(4):339–350CrossRefGoogle Scholar
  3. 3.
    Bauer S, Clark D, Lehr W (2010) Understanding broadband speed measurements. Social Science Electronic PublishingGoogle Scholar
  4. 4.
    China Internet Network Information Center (2017) Statistical report on internet development in china, 2016. Technical reportGoogle Scholar
  5. 5.
    Dobrian F, Sekar V, Awan A, Stoica I, Joseph D, Ganjam A, Zhan J, Zhang H (2011) Understanding the impact of video quality on user engagement. Proc ACM SIGCOMM 2011:362–373CrossRefGoogle Scholar
  6. 6.
    Finamore A, Mellia M, Torres R, Rao SG (2011) Youtube everywhere: impact of device and infrastructure synergies on user experience. ACM SIGCOMM Conf Internet Measure Conf: 345–360Google Scholar
  7. 7.
    Huang Y, Fu TZ, Chiu D-M, Lui JC, Huang C (2008) Challenges, design and analysis of a large-scale p2p-vod system. Acm Sigcomm Comput Commun Rev 38(4):375–388CrossRefGoogle Scholar
  8. 8.
    (2013) Internet connection speed recommendations.
  9. 9.
    Jiang J, Sekar V, Stoica I, Zhang H (2013) Shedding light on the structure of internet video quality problems in the wild. ACM Conf Emerg Netwo Experiments Technol: 357–368Google Scholar
  10. 10.
    Kanrar S, Mandal NK (2016) Video traffic flow analysis in distributed system during interactive session. Adv Multimed 2016(1):1–14Google Scholar
  11. 11.
    Kim Y, Steiner P (2016) Quasi-experimental designs for causal inference. Educ Psychol 51(3–4):395–405CrossRefGoogle Scholar
  12. 12.
    Krishnan SS, Sitaraman RK (2013) Video stream quality impacts viewer behavior: inferring causality using quasi-experimental designs. IEEE/ACM Trans Networking 21(6):2001–2014CrossRefGoogle Scholar
  13. 13.
    Kullback S, Leibler RA (1951) On information and sufficiency. Ann Math Stat 22(1):79–86MathSciNetCrossRefGoogle Scholar
  14. 14.
    Li Z, Lin J, Akodjenou MI, Xie G, Peng G, Peng G, et al. (2012). Watching videos from everywhere: a study of the PPTV mobile VoD system. ACM Conf Internet Measure Conf: 185–198Google Scholar
  15. 15.
    Li Z, Xie G, Lin J, Jin Y, Kaafar MA, Salamatian K (2014). On the geographic patterns of a large-scale mobile video-on-demand system. Proc IEEE Infocom: 397–405Google Scholar
  16. 16.
    Li Z, Wu Q, Salamatian K, Xie G (2015) Video delivery performance of a large-scale vod system and the implications on content delivery. IEEE Trans Multimed 17(6):880–892CrossRefGoogle Scholar
  17. 17.
    Li Z, Xie G, Kaafar MA, Salamatian K (2015) User behavior characterization of a large-scale Mobile live streaming system. Proc 24th Int Conf World Wide Web: 307–313Google Scholar
  18. 18.
    Liu X, Dobrian F, Milner H, Jiang J, Sekar V, Stoica I, et al (2012) A case for a coordinated internet video control plane. Proc ACM SIGCOMM: 359–370Google Scholar
  19. 19.
    Mahimkar AA, Ge Z, Shaikh A, Wang J, Yates J, Zhang Y et al (2009) Towards automated performance diagnosis in a large iptv network. Acm Sigcomm Comput Commun Rev 39(4):231–242CrossRefGoogle Scholar
  20. 20.
    Mitchell TM (2003) Machine learning. China Machine PressGoogle Scholar
  21. 21.
    Mu M, Mauthe A, Garcia F (2008) A utility-based QoS model for emerging multimedia applications. Proc 2nd Int Conf Next Gen Mobile Appl Serv Technol NGMAST 2008, pp 521–528Google Scholar
  22. 22.
    Nam H, Kim KH, Schulzrinne H (2016) QoE matters more than QoS: why people stop watching cat videos. 35th Ann IEEE Int Conf Comput Commun IEEE INFOCOM 2016:1–9Google Scholar
  23. 23.
    Otto JS, Choffnes DR, Siganos G (2011) On blind mice and the elephant: understanding the network impact of a large distributed system. Proc ACM SIGCOMM: 110–121Google Scholar
  24. 24.
    Rao A, Legout A, Lim YS, Towsley D, Barakat C, Dabbous W (2011) Network characteristics of video streaming traffic. Proc Conf Emerg Netw Experiments Technol: 25Google Scholar
  25. 25.
    Shafiq MZ, Erman J, Ji L, Liu AX, Pang J, Wang J (2014) Understanding the impact of network dynamics on mobile video user engagement. Acm Sigmetrics Perform Eval Rev 42(1):367–379CrossRefGoogle Scholar
  26. 26.
    Stuart A, Ord K, Arnold S (2008) Kendall's advanced theory of statistics; volume 2A, classical inference and the linear model. Edward Arnold, LondonGoogle Scholar
  27. 27.
    Team cymru (2013)
  28. 28.
    University of oregon route views project.
  29. 29.
    Yu M, Jiang W, Li H, Stoica I (2012) Tradeoffs in CDN designs for throughput oriented traffic. Int Conf Emerg Netw Experiments Technol: 145–156Google Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of ComputerCity College of Dongguan University of TechnologyDongguanChina

Personalised recommendations