Cross-Layer Effects on Training Neural Algorithms for Video Streaming

ABSTRACT

Nowadays Dynamic Adaptive Streaming over HTTP (DASH) is the most prevalent solution on the Internet for multimedia streaming and responsible for the majority of global traffic. DASH uses adaptive bit rate (ABR) algorithms, which select the video quality considering performance metrics such as throughput and playout buffer level. Pensieve is a system that allows to train ABR algorithms using reinforcement learning within a simulated network environment and is outperforming existing approaches in terms of achieved performance. In this paper, we demonstrate that the performance of the trained ABR algorithms depends on the implementation of the simulated environment used to train the neural network. We also show that the used congestion control algorithm impacts the algorithms' performance due to cross-layer effects.

References

1. Claudio Alberti, Daniele Renzi, Christian Timmerer, and Christopher Mueller. Automated QoE Evaluation of Dynamic Adaptive Streaming Over HTTP. 5th International Workshop on Quality of Multimedia Experience (QoMEX), 2013.
2. M. Allman, V. Paxson, and E. Blanton. TCP Congestion Control. RFC 5681 (Draft Standard), September 2009.
3. Divyashri Bhat, Amr Rizk, and Michael Zink. Not so QUIC: A Performance Study of DASH over QUIC. In Proceedings of the 27th Workshop on Network and Operating Systems Support for Digital Audio and Video, pages 13--18. ACM, 2017.
4. Neal Cardwell, Yuchung Cheng, C Stephen Gunn, Soheil Hassas Yeganeh, and Van Jacobson. BBR: Congestion-Based Congestion Control. ACM Queue, 14(5):50:20--50:53, 2016.
5. Cisco. The Zettabyte Era: Trends and Analysis. (June 2017):1--32, 2017.
6. Jairo Esteban, Steven Benno, Andre Beck, Yang Guo, Volker Hilt, and Ivica Rimac. Interactions between HTTP adaptive streaming and TCP. Proceedings of the 22nd international workshop on Network and Operating System Support for Digital Audio and Video - NOSSDAV '12, pages 21--26, 2012.
7. Sangtae Ha, Injong Rhee, and Lisong Xu. CUBIC: A New TCP-Friendly HighSpeed TCP Variant. ACM SIGOPS Operating Systems Review - Research and developments in the Linux kernel, 42(5):64--74, 2008.
8. Te-Yuan Huang, Nikhil Handigol, Brandon Heller, Nick NcKeown, and Ramesh Johari. Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard. Proceedings of the ACM conference on Internet measurement, pages 225--238, 2012.
9. Yao Liu, Sujit Dey, Fatih Ulupinar, Michael Luby, and Yinian Mao. Deriving and Validating User Experience Model for DASH Video Streaming. IEEE Transactions on Broadcasting, 61(4):651--665, 2015.
10. Hongzi Mao, Ravi Netravali, and Mohammad Alizadeh. Neural Adaptive Video Streaming with Pensieve. In Proceedings of the Conference of the ACM Special Interest Group on Data Communication, pages 197--210. ACM, 2017.
11. Volodymyr Mnih, Adria Puigdomenech Badia, Mehdi Mirza, Alex Graves, Timothy Lillicrap, Tim Harley, David Silver, and Koray Kavukcuoglu. Asynchronous methods for deep reinforcement learning. In International Conference on Machine Learning, pages 1928--1937, 2016.
12. Yongtao Shuai and Thorsten Herfet. On stabilizing buffer dynamics for adaptive video streaming with a small buffering delay. In 14th IEEE Annual Consumer Communications & Networking Conference (CCNC), pages 435--440. IEEE, 2017.
13. Kevin Spiteri, Rahul Urgaonkar, and Ramesh K. Sitaraman. BOLA: Near-optimal bitrate adaptation for online videos. In Proceedings of the 35th Annual IEEE International Conference on Computer Communications, IEEE INFOCOM, 2016.
14. Denny Stohr, Alexander Frömmgen, Jan Fornoff, Michael Zink, Alejandro Buchmann, and Wolfgang Effelsberg. QoE Analysis of DASH Cross-Layer Dependencies by Extensive Network Emulation. In Proceedings of the workshop on QoE-based Analysis and Management of Data Communication Networks, 2016.
15. Christian Timmerer and Alan Bertoni. Advanced Transport Options for the Dynamic Adaptive Streaming over HTTP. arXiv preprint arXiv:1606.00264, 2016.
16. Cong Wang, Amr Rizk, and Michael Zink. SQUAD: a spectrum-based quality adaptation for dynamic adaptive streaming over HTTP. In Proceedings of the 7th International Conference on Multimedia Systems, MMSys 2016, pages 1 -- 12, 2016.
17. Xiaoqi Yin, Abhishek Jindal, Vyas Sekar, and Bruno Sinopoli. A Control-Theoretic Approach for Dynamic Adaptive Video Streaming over HTTP. Proceedings of the 2015 ACM SIGCOMM, pages 325--338, 2015.

Index Terms

• Information systems

  • Information systems applications

    • Multimedia information systems

      • Multimedia streaming

• Networks

  • Network algorithms

    • Control path algorithms

      • Network control algorithms

  • Network performance evaluation

    • Network experimentation

    • Network measurement