A brief history of streaming video in the Internet
1. A Brief History of
Streaming Video in the
Internet
Lessons Learned and Future Directions
ARS Lab - Carleton University
May 2015
Gabriel Wainer / Stenio Fernandes
2. The lost decades (1/2)
● 70’s - not enough network capacity to
support multimedia
● 80’s and early 90’s- researchers started to
investigate ATM network support for audio
and video
○ This is “kind of” link+network layer technology
○ It supports ABR, VBR, and CBR
○ Investigation on transport protocols
3. The lost decades (2/2)
● 90’s - Improvements on audio and video
CODECs
○ ATM was going nowhere
○ IP layer-only support would be the key
■ Discussion on Diffserv and Intserv started
○ Some applications have arisen
■ Yahoo Messenger, Microsoft MSN, IP Telephony
4. An interesting decade (1/3)
● early to mid 2000’s
○ Server side focus
■ Rate adaptation (rate control and shaping)
● CODEC level
● Network or Transport level
○ Proxy Caching / IP Multicasting
○ Multiple video quality levels were unacceptable
■ Waste of storage resources
○ Killer application
■ Skype
5. An interesting decade (2/3)
● mid to late 2000’s
○ Real and fast advances on CODECs
■ Scalable Video (Multiple layers)
■ Error Concealment Techniques
○ Network with enough capacity (at the edge)
○ Killer applications
■ YouTube, Web Conferencing
■ P2P Streaming
6. An interesting decade (3/3)
● mid to late 2000’s
○ Design of Transport Protocols
■ Beyond the RTP Stack (RCTP, RSTP, RTP)
● TCP Friendly Rate Control (TFRC)
● Datagram Congestion Control Protocol (DCCP)
● eXplicit Control Protocol (XCP)
○ Support from Content Delivery Networks (CDN)
■ Akamai
○ Still ONE long video file to be delivered (for VoD)
7. The Current State (1/2)
● 2010’s
○ Lots of Pressure for Video Content
○ Movie and Broadcast industry finally jumped into
this new model
■ Hulu, Netflix, HBO
○ Wireless Networks
with enough capacity
■ Expensive though
■ Prioritization of packets
● new 802.11, WiMAX, and 3G-5G strategies
8. The Current State (2/2)
● 2010’s
○ Storage is not really an actual problem
■ Thanks to cloud computing services
○ So, multiple video qualities are now feasible
○ Middleboxes are still a problem
■ Blocking and throttling traffic
■ Security features
○ So, the solution would be deliver multimedia
streaming over HTTP
9. DASH was born
● Some issues before the design of DASH
○ But, TCP is bad for delivering multimedia content
with strict playout time
■ Due to the intrinsic congestion control and
reliability mechanisms
○ And UDP will kill the network
■ May leading to congestion collapse
○ And servers (even in a cloud environment) may not
handle flash crowds
10. Techniques for Adaptive Streaming
T. Stockhammer: “Dynamic Adaptive Streaming over HTTP-Design Priciples and Standards” In: MMSys ’11: Proceedings of the
second annual ACM conference on Multimedia systems New York, NY, USA: ACM Press , Feb 2011, S. 133-144
11. Techniques for Adaptive Streaming
Luca De Cicco, Saverio Mascolo, and Vittorio Palmisano. 2011. Feedback control for adaptive live video
streaming. In Proceedings of the second annual ACM conference on Multimedia systems(MMSys '11)
80’s and 90’s From 90’s to mid 2000’s From mid 2000’s to now
12. DASH was born (2/2)
● Dynamic Adaptive Streaming over HTTP
○ Give the clients the power to decide on the
“quality” they want to receive
■ Given the network conditions
■ In real time
○ HTTP over TCP will pass through most middleboxes
■ Use small (but not too small) chunks to “avoid”
TCP behaviour “issues”
○ Adaptation Logic is open for investigation
15. Control-Theoretic Approaches for the
Adaptation Logic - Examples (1/5)
Hsiao YM, Chen CH, Lee JF, Chu
YS. "Designing and implementing a
scalable video-streaming system
using an adaptive control scheme,"
Consumer Electronics, IEEE
Transactions on , vol.58, no.4, pp.
1314,1322, November 2012
16. Control-Theoretic Approaches for the
Adaptation Logic - Examples (2/5)
De Cicco, L.; Caldaralo, V.; Palmisano, V.; Mascolo, S., "ELASTIC: A Client-Side Controller for Dynamic Adaptive
Streaming over HTTP (DASH)," Packet Video Workshop (PV), 2013 20th International , vol., no., pp.1,8, 12-13 Dec. 2013
17. Control-Theoretic Approaches for the
Adaptation Logic - Examples (3/5)
Luca De Cicco, Saverio Mascolo, and Vittorio Palmisano. 2011. Feedback control for adaptive live video
streaming. In Proceedings of the second annual ACM conference on Multimedia systems(MMSys '11)
18. Control-Theoretic Approaches for the
Adaptation Logic - Examples (4/5)
Fortuna R, Grieco LA, Boggia G, Camarda P. “Quality adaptive end-to-end packet scheduling to avoid playout interruptions in
Internet video streaming systems,” Journal of Systems and Software, Volume 83, Issue 8, August 2010, Pages 1489-1499
19. Control-Theoretic Approaches for the
Adaptation Logic - Examples (5/5)
Ito, M., Fernandes, S., et al, “A Fine-Tuned Control-Theoretic Approach for Dynamic Adaptive Streaming Over HTTP”, IEEE
International Symposium on Computers and Communications (IEEE ISCC 2015),, Lanarca, Cyprus, 2015
20. Control-Theoretic Approaches for the
Adaptation Logic - Examples (5/5)
Ito, M., Fernandes, S., et al, “A Fine-Tuned Control-Theoretic Approach for Dynamic Adaptive Streaming Over HTTP”, IEEE International
Symposium on Computers and Communications (IEEE ISCC 2015),, Lanarca, Cyprus, 2015
Two components: Yet Another Control System + State Machine
State Machine: A pure control system is sufficient for buffer stabilization, but it does not ensure
playout smoothness.
21. Control-Theoretic Approaches for the
Adaptation Logic - Examples (5/5)
Ito, M., Fernandes, S., et al, “A Fine-Tuned Control-Theoretic Approach for Dynamic Adaptive Streaming Over HTTP”, IEEE International
Symposium on Computers and Communications (IEEE ISCC 2015),, Lanarca, Cyprus, 2015
23. Research Challenges (1/2)
● Adaptation Logic
○ Include raw factors
■ Application Level: # of stalls
■ Transport Level: Buffer utilization
■ Network Level: Packet losses and delays
○ More precise control
■ Optimization strategies
■ Control-Theoretic Approaches
■ Markov-Modulated Decision Process
24. Research Challenges (2/2)
● New Networking Scenarios
○ SDN and NFV
■ support from OR
■ On top of
○ LTE to 5G Environments
● Include Quality of Experience (QoE) Models in the
control loop
● P2P-like Strategies for Scalable DASH Systems in
Wireless Environments
25. Some references
1. Hsiao YM, Chen CH, Lee JF, Chu YS. "Designing and implementing a scalable video-streaming system using an adaptive
control scheme," Consumer Electronics, IEEE Transactions on , vol.58, no.4, pp.1314,1322, November 2012
2. Cicco L, Mascolo S, "An Adaptive Video Streaming Control System: Modeling, Validation, and Performance Evaluation,"
Networking, IEEE/ACM Transactions on , vol.22, no.2, pp.526,539, April 2014
3. Changuel, N.; Sayadi, B.; Kieffer, M., "Control of Multiple Remote Servers for Quality-Fair Delivery of Multimedia
Contents," Selected Areas in Communications, IEEE Journal on , vol.32, no.4, pp.746,759, April 2014
4. Ito MS, Antonello R, Sadok D, Fernandes S. “Network Level Characterization of Adaptive Streaming over HTTP
Applications”, In IEEE Symposium on Computers and Communications (ISCC’14)
5. Jiang J, Sekar V, Zhang H. “Improving fairness, efficiency, and stability in HTTP-based adaptive video streaming with
FESTIVE”. In Proceedings of the 8th international conference on Emerging networking experiments and technologies
(CoNEXT '12)
6. Li B, Wang Z, Liu J, Zhu W. “Two decades of internet video streaming: A retrospective view”. ACM Trans. Multimedia
Comput. Commun. Appl. 9, 1s, Article 33 (October 2013)
7. Li Z, Zhu X, Gahm J, Pan R, Hu H, Begen AC, Oran D. "Probe and Adapt: Rate Adaptation for HTTP Video Streaming At
Scale," Selected Areas in Communications, IEEE Journal on , vol.32, no.4, pp.719,733, April 2014
8. Patras P, Banchs A, Serrano P. “A control theoretic scheme for efficient video transmission over IEEE 802.11e EDCA
WLANs”. ACM Trans. Multimedia Comput. Commun. Appl. 8, 3, Article 29 (August 2012)
9. Zhou C, Lin CW, Zhang X, Guo Z. "A Control-Theoretic Approach to Rate Adaption for DASH Over Multiple Content
Distribution Servers," Circuits and Systems for Video Technology, IEEE Transactions on , vol.24, no.4, pp.681,694, 2014
10. Yingsong Huang; Shiwen Mao; Midkiff, S.F., "A Control-Theoretic Approach to Rate Control for Streaming Videos,"
ultimedia, IEEE Transactions on , vol.11, no.6, pp.1072,1081, Oct. 2009