Modeling large-scale live video streaming client behavior

The live streaming of large-scale events over the Internet attracts a highly diverse audience. Despite the progress in streaming technologies, which employ adaptive bitrate to dynamically adjust the streaming of individual clients to reflect their resource availability, content providers still struggle to efficiently match provisioned capacity and demand in such events. In this paper, we present an in-depth characterization and modeling of client behavior during the live streaming of the 2018 FIFA World Soccer Cup. We analyze logs covering more than 60 million streaming sessions collected from servers of one of the major content providers in Latin America. We characterize key features of the workload, such as the number and duration of sessions as well as transmission quality (i.e., bitrate), shedding new light on the workload of a major content provider during a unique large-scale streaming event. We then propose a simple hierarchical model to describe the typical behavior of individual clients both at the session and video bitrate adaptation layers. Taking a step further, we employ non-supervised clustering to identify classes of client behavior and generate specialized models, one for each class. Our evaluation shows that the specialized models are more accurate compared to the single general model, as it can better describe the diversity of client behavior patterns.

[1]  Michael Seufert,et al.  Scoring High: Analysis and Prediction of Viewer Behavior and Engagement in the Context of 2018 FIFA WC Live Streaming , 2020, ACM Multimedia.

[2]  Amin Vahdat,et al.  MediSyn: a synthetic streaming media service workload generator , 2003, NOSSDAV '03.

[3]  Jerome A. Rolia,et al.  A Synthetic Workload Generation Technique for Stress Testing Session-Based Systems , 2006, IEEE Transactions on Software Engineering.

[4]  Mariacarla Calzarossa,et al.  Workload Characterization , 2016, ACM Comput. Surv..

[5]  Ramesh K. Sitaraman,et al.  Video Stream Quality Impacts Viewer Behavior: Inferring Causality Using Quasi-Experimental Designs , 2012, IEEE/ACM Transactions on Networking.

[6]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[7]  Alex Borges Vieira,et al.  Characterizing SopCast client behavior , 2012, Comput. Commun..

[8]  Mostafa H. Ammar,et al.  Prefix-preserving IP address anonymization: measurement-based security evaluation and a new cryptography-based scheme , 2004, Comput. Networks.

[9]  Jaideep Chandrashekar,et al.  Characterizing Client Behavior of Commercial Mobile Video Streaming Services , 2014, MoVid@MMSys.

[10]  Johan Tordsson,et al.  Analysis and characterization of a video-on-demand service workload , 2015, MMSys.

[11]  Keith W. Ross,et al.  A Measurement Study of a Large-Scale P2P IPTV System , 2007, IEEE Transactions on Multimedia.

[12]  Ramesh K. Sitaraman,et al.  Video Stream Quality Impacts Viewer Behavior: Inferring Causality Using Quasi-Experimental Designs , 2013, IEEE/ACM Transactions on Networking.

[13]  Jun Liu,et al.  Large-Scale User Behavior Characterization of Online Video Service in Cellular Network , 2016, IEEE Access.

[14]  Danilo Giordano,et al.  Five Years at the Edge: Watching Internet From the ISP Network , 2018, IEEE/ACM Transactions on Networking.

[15]  Anirban Mahanti The Evolving Streaming Media Landscape , 2014, IEEE Internet Comput..

[16]  Yong Liu,et al.  Measurement and Modeling of Video Watching Time in a Large-Scale Internet Video-on-Demand System , 2013, IEEE Transactions on Multimedia.

[17]  Koenraad Laevens,et al.  Session reconstruction for HTTP adaptive streaming: Laying the foundation for network-based QoE monitoring , 2012, 2012 IEEE 20th International Workshop on Quality of Service.

[18]  E. Forgy,et al.  Cluster analysis of multivariate data : efficiency versus interpretability of classifications , 1965 .

[19]  Srinivasan Seshan,et al.  Developing a predictive model of quality of experience for internet video , 2013, SIGCOMM.

[20]  Larry Wasserman,et al.  All of Statistics: A Concise Course in Statistical Inference , 2004 .

[21]  Virgílio A. F. Almeida,et al.  A hierarchical characterization of a live streaming media workload , 2006 .

[22]  Gaogang Xie,et al.  User Behavior Characterization of a Large-scale Mobile Live Streaming System , 2015, WWW.

[23]  M. Stephens EDF Statistics for Goodness of Fit and Some Comparisons , 1974 .

[24]  R. L. Thorndike Who belongs in the family? , 1953 .

[25]  Ítalo S. Cunha,et al.  Characterizing QoE in Large-Scale Live Streaming , 2017, GLOBECOM 2017 - 2017 IEEE Global Communications Conference.

[26]  Ítalo S. Cunha,et al.  Internet Performance from Facebook's Edge , 2019, Internet Measurement Conference.

[27]  SeshanSrinivasan,et al.  Developing a predictive model of quality of experience for internet video , 2013 .

[28]  Rocky K. C. Chang,et al.  Measuring the quality of experience of HTTP video streaming , 2011, 12th IFIP/IEEE International Symposium on Integrated Network Management (IM 2011) and Workshops.

[29]  Gaogang Xie,et al.  Characterizing and Modeling User Behavior in a Large-Scale Mobile Live Streaming System , 2017, IEEE Transactions on Circuits and Systems for Video Technology.

[30]  Alan C. Bovik,et al.  Study of the effects of stalling events on the quality of experience of mobile streaming videos , 2014, 2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[31]  Flavio Figueiredo,et al.  On the Dynamics of Social Media Popularity: A YouTube Case Study , 2014, TOIT.

[32]  Anne-Marie Kermarrec,et al.  Large scale analysis of HTTP Adaptive Streaming in mobile networks , 2013, 2013 IEEE 14th International Symposium on "A World of Wireless, Mobile and Multimedia Networks" (WoWMoM).

[33]  Chao Gui,et al.  Content-aware adaptation scheme for QoE optimized dash applications , 2014, 2014 IEEE Global Communications Conference.

[34]  Alex Borges Vieira,et al.  Workload models and performance evaluation of cloud storage services , 2016, Comput. Networks.

[35]  Azer Bestavros,et al.  GISMO: a Generator of Internet Streaming Media Objects and workloads , 2001, PERV.

[36]  D. Brillinger,et al.  Handbook of methods of applied statistics , 1967 .

[37]  Markus Fiedler,et al.  Controlled vs. Uncontrolled Degradations of QoE : The Provisioning-Delivery Hysteresis in Case of Video , 2010 .

[38]  Laurent Vanbever,et al.  Scalable programmable inbound traffic engineering , 2015, SOSR.

[39]  Carey L. Williamson,et al.  ProWGen: a synthetic workload generation tool for simulation evaluation of web proxy caches , 2002, Comput. Networks.

[40]  Understanding the impact of video quality on user engagement , 2011, SIGCOMM.

[41]  Dina Katabi,et al.  A framework for scalable global IP-anycast (GIA) , 2000, SIGCOMM 2000.