Abstract
Geosynchronous satellite (GEO) networks are an important Internet access option for users beyond terrestrial connectivity. However, unlike terrestrial networks, GEO networks exhibit high latency and deploy TCP proxies and traffic shapers. The deployment of proxies effectively mitigates the impact of high network latency in GEO networks, while traffic shapers help realize customer-controlled data-saver options that optimize data usage. However, it is unclear how the interplay between GEO networks’ high latency, TCP proxies, and traffic-shaping policies affects the quality of experience for commonly used video applications. To address this gap, we analyze the quality of over 2 k YouTube video sessions streamed across a production GEO network with a 900 Kbps shaping rate. Given the average bit rates of the videos, we expected streaming to be seamless at resolutions of 360p, and nearly seamless at resolutions approaching 480p. However, our analysis reveals that this is not the case: \(30\%\) of both TCP and QUIC sessions experience rebuffering, while the median average resolution is only 404p for TCP and 360p for QUIC. Our analysis identifies two key factors that contribute to sub-optimal performance: (i) unlike TCP, QUIC only utilizes \(70\%\) of the network capacity; and (ii) YouTube’s chunk request pipelining neglects network latency, resulting in idle periods that disproportionately harm the throughput of smaller chunks. As a result of our study, Viasat discontinued support for the low-bandwidth data-saving option in U.S. business and residential markets to avoid potential degradation of video quality—highlighting the practical significance of our findings.
Udit Paul was a PhD student at the University of California, Santa Barbara when the work was performed.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
For instance, one study states that, in 2022, YouTube represented \(15\%\) of traffic on consumer broadband networks, while Netflix represented \(9\%\) [6].
- 2.
Note that traffic shaping is a subscriber opt-in feature for the ISP in the study.
- 3.
Categories collected are: Sports, Education, Science & Technology, Shows, Pets & Animals, Nonprofits & Activism, News & Politics, Gaming, Music, Comedy, People & Blogs, Autos & Vehicles, Film & Animation, Entertainment, Howto & Style, Travel & Events. Categories such as Sports are usually of higher bit rate compared to Education.
References
BBR development group. https://groups.google.com/g/bbr-dev, Accessed 13 Jan 2024
Abu-El-Haija, S., Kothari, N., Lee, J., Natsev, P., Toderici, G., Varadarajan, B., Vijayanarasimhan, S.: Youtube-8M: a large-scale video classification benchmark. arXiv preprint arXiv:1609.08675 (2016)
Adhikari, V.K., Jain, S., Chen, Y., Zhang, Z.L.: Vivisecting YouTube: an active measurement study. In: IEEE INFOCOM 2012 (2012)
Bhat, D., Rizk, A., Zink, M.: Not so QUIC: a performance study of DASH over QUIC, NOSSDAV 2017, pp. 13–18 (2017)
Border, J., Shah, B., Su, C.J., Torres, R.: Evaluating QUIC’s performance against performance enhancing proxy over satellite link. In: IFIP Networking Conference, pp. 755–760 (2020)
Cantor, L.: The Global Internet Phenomena Report. Technical Report, Sandvine, Waterloo, ON, Canada (2022)
Flach, T., et al.: An internet-wide analysis of traffic policing, SIGCOMM 2016, pp. 468–482. ACM (2016)
Ge, C., et al.: QoE-assured live streaming via satellite backhaul in 5G networks. IEEE Trans. Broadcast. 65, 381–391 (2019)
Gutterman, C., et al.: Requet: Real-Time QoE detection for encrypted YouTube traffic, MMSys 2019, pp. 48–59. ACM (2019)
Hoßfeld, T., Seufert, M., Hirth, M., Zinner, T., Tran-Gia, P., Schatz, R.: Quantification of YouTube QoE via Crowdsourcing. In: IEEE International Symposium on Multimedia, pp. 494–499 (2011)
Huang, T.Y., Ekanadham, C., Berglund, A.J., Li, Z.: Hindsight: evaluate video bitrate adaptation at scale, ACM MMSys 2019, pp. 86–97 (2019)
Khokhar, M.J., Ehlinger, T., Barakat, C.: From network traffic measurements to QoE for internet video. In: IFIP Networking Conference (2019)
Kuhn, N., Michel, F., Thomas, L., Dubois, E., Lochin, E.: QUIC: opportunities and threats in SATCOM. In: ASMS/SPSC (2020)
Langley, A., et al.: The QUIC transport protocol: design and internet-scale deployment, ACM SIGCOMM 2017, pp. 183–196 (2017)
Li, F., Niaki, A.A., Choffnes, D., Gill, P., Mislove, A.: A large-scale analysis of deployed traffic differentiation practices, ACM SIGCOMM 2019, pp. 130–144 (2019)
Lv, G., Wu, Q., Wang, W., Li, Z., Xie, G.: Lumos: towards better video streaming QOE through accurate throughput prediction. In: IEEE INFOCOM 2022, pp. 650–659 (2022)
Mansy, A., Ammar, M., Chandrashekar, J., Sheth, A.: Characterizing client behavior of commercial mobile video streaming services. In: ACM MoViD 2014 (2018)
Mao, H., Netravali, R., Alizadeh, M.: Neural adaptive video streaming with pensieve, SIGCOMM 2017, pp. 197–210. ACM (2017)
Joras, M., Chi, Y.: How Facebook is bringing QUIC to billions. https://engineering.fb.com/2020/10/21/networking-traffic/how-facebook-is-bringing-quic-to-billions
Megyesi, P., Krämer, Z., Molnár, S.: How quick is QUIC? In: 2016 IEEE ICC (2016)
Mok, R.K., Chan, E.W., Luo, X., Chang, R.K.: Inferring the QoE of HTTP Video streaming from user-viewing activities, W-MUST 2011, pp. 31–36. ACM (2011)
Mondal, A., et al.: Candid with YouTube: adaptive streaming behavior and implications on data consumption, NOSSDAV 2017, pp. 19–24. ACM (2017)
Nam, H., Kim, K.H., Calin, D., Schulzrinne, H.: YouSlow: a performance analysis tool for adaptive bitrate video streaming. SIGCOMM Comput. Commun. Rev. 44, 111–112 (2014)
Nam, H., Kim, K.H., Schulzrinne, H.: QoE matters more than QoS: why people stop watching cat videos. In: INFOCOM 2016, IEEE (2016)
Nam, Y.S., et al.: Xatu: richer neural network based prediction for video streaming. Meas. Anal. Comput. Syst. 5(3), 1–26 (2021)
Ramachandran, S., Gryta, T., Dapena, K., Thomas, P.: The truth about faster internet: It’s not worth it. Wall Street J. (2019). https://www.wsj.com/graphics/faster-internet-not-worth-it/
Raman, A., Varvello, M., Chang, H., Sastry, N., Zaki, Y.: Dissecting the performance of satellite network operators. In: CoNEXT 2023, ACM (2023)
Reznik, S., Reut, D., Shustilova, M.: Comparison of geostationary and low-orbit “round dance” satellite communication systems. In: IOP Conference Series: Materials Science and Engineering, vol. 971 (2020)
Rüth, J., Wolsing, K., Wehrle, K., Hohlfeld, O.: Perceiving QUIC: do users notice or even care?, CoNEXT 2019, pp. 144–150. ACM (2019)
Seufert, M., Egger, S., Slanina, M., Zinner, T., Hoßfeld, T., Tran-Gia, P.: A survey on quality of experience of HTTP adaptive streaming. IEEE Commun. Surv. Tutorials 17, 469–492 (2015)
Seufert, M., Schatz, R., Wehner, N., Casas, P.: QUICker or not? an empirical analysis of QUIC vs TCP for video streaming QoE provisioning. In: 2019 ICIN, pp. 7–12 (2019)
Seufert, M., Schatz, R., Wehner, N., Gardlo, B., Casas, P.: Is QUIC becoming the new TCP? On the potential impact of a new protocol on networked multimedia QoE. In: 2019 Eleventh International Conference on Quality of Multimedia Experience (QoMEX) (2019)
Spang, B., et al.: Sammy: Smoothing video traffic to be a friendly internet neighbor, SIGCOMM 2023, pp. 754–768. ACM (2023)
Thibaud, A., Fasson, J., Arnal, F., Pradas, D., Dubois, E., Chaput, E.: QoE enhancements on satellite networks through the use of caches. Int. J. Satell. Commun. Network. 36, 553–565 (2018)
Thomas, L., Dubois, E., Kuhn, N., Lochin, E.: Google QUIC performance over a public SATCOM access. Int. J. Satell. Commun. Network. 37, 601–611 (2019)
Wamser, F., Seufert, M., Casas, P., Irmer, R., Tran-Gia, P., Schatz, R.: YoMoApp: a tool for analyzing QoE of YouTube HTTP adaptive streaming in mobile networks. In: EuCNC 2015, pp. 239–243 (2015)
Xu, S., Wang, X., Huang, M.: Modular and deep QoE/QoS mapping for multimedia services over satellite networks. Int. J. Commun. Syst. 31, e3793 (2018)
Yin, X., Jindal, A., Sekar, V., Sinopoli, B.: A control-theoretic approach for dynamic adaptive video streaming over HTTP. In: SIGCOMM Computer Communication Review, pp. 325–338 (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
A Appendix
A Appendix
1.1 A.1 Ethical Considerations
Although our work involves HTTP log analysis on an operational GEO satellite network, our work is not human subjects research. At no point is any data collected from the customers of the network. We collect and analyze only our own experimentally generated traffic.
Idle time of campus network.
Post 30 s throughput without considering idle time. Vertical line is 900 Kbps.
Chunk size vs throughput over TCP.
Achieved throughput. Vertical line is 900 Kbps.
Time to first byte of each chunk. Vertical line is 600 ms.
1.2 A.2 Supplementary Results
In this section we include some additional, supplementary graphs that were briefly described in the main body of the paper. The median idle time for both TCP and QUIC was short in our campus network experiment, around 15 ms, as shown in Fig. 12. This result suggests that the pipelining inefficiency is magnified by the high round trip time of the GEO satellite network. Figure 13 shows the \(T_{network}\) after 30 seconds of playback, in order to eliminate any effect due to slow start. The figure indicates that QUIC throughput still varies well below the shaped bandwidth 900 kbps. This indicates that congestion control, and specifically the initial slow start, are not the source of the low throughput. Figure 16 shows the TTFB of each chunk. We can observe that almost all chunks have a TTFB larger than 600 ms; QUIC in particular forms a cluster close to 600 ms. The correlation between achieved throughput (\(T_{idle}\)) and chunk size for TCP is illustrated in Fig. 14. The Pearson statistic for correlation of achieved throughput and log(chunk size) is 0.62. Finally, Fig. 15 shows that the \(T_{idle}\) of TCP outperforms that of QUIC in GEO networks; the median TCP throughput is 0.58 Mbps, while QUIC’s median throughput is 0.47 Mbps. Importantly, however, neither reach the shaped bandwidth rate. Figure 16 shows the TTFB of each chunk. The median chunk TTFB for TCP is 1.21 s, while it is 0.78 s for QUIC.
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Liu, J., Lerner, D., Chung, J., Paul, U., Gupta, A., Belding, E. (2024). Watching Stars in Pixels: The Interplay Of Traffic Shaping and YouTube Streaming QoE over GEO Satellite Networks. In: Richter, P., Bajpai, V., Carisimo, E. (eds) Passive and Active Measurement. PAM 2024. Lecture Notes in Computer Science, vol 14538. Springer, Cham. https://doi.org/10.1007/978-3-031-56252-5_8
Download citation
DOI: https://doi.org/10.1007/978-3-031-56252-5_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-56251-8
Online ISBN: 978-3-031-56252-5
eBook Packages: Computer ScienceComputer Science (R0)