Congestion-Aware Video Streaming over Peer-to-Peer Networks Eric Setton Information Systems Laboratory Stanford University

2 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Multimedia Distribution over the Internet BitTorrent, Kazaa, Napster Self-scaling, peers relay data Typical latency: a few hours ~ a few days P2P file transfer network Seed peer Content delivery network Media server(s).. Content providers, Akamai Over-provisioned infrastructure e.g servers deployed by AOL for Live 8 concert in July 2005 Typical latency: a few seconds

3 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Live Multicast over Peer-to-Peer Challenges Limited bandwidth Delay due to multi-hop transmission Dynamic behavior of peers Related work Coopnet [Padmanabhan, Wang and Chou, 2003] Coolstreaming [Zhang, Liu, Li and Yum, 2005] Approach Determine encoding rate as a function of network bandwidth Adapt media scheduling to network conditions and to content Build and maintain multicast trees, request retransmissions to mitigate losses … … Video stream

4 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Distortion model for throughput-limited streaming –Server-client scenario –Extension to peer-to-peer Packet scheduling for low-latency video streaming –Congestion-distortion optimized (CoDiO) scheduling –Low complexity (CoDiO light) scheduler Peer-to-peer video streaming –Distributed peer-to-peer protocol –Receiver-driven CoDiO light retransmission requests –Sender-driven CoDiO light prioritization Presentation Outline

5 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Video Distortion with Self Congestion Good picture quality Bad picture quality Bit-rate (kb/s) Self congestion causes late loss

6 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Modeling Streaming Performance PSNR (dB) P loss Bit-rate (kb/s) [ Stuhlmüller, Färber, Link and Girod, 2000 ]

7 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Influence of T and K ForemanSalesman Simulations over ns-2 Bandwidth 400 kb/s Simulations over ns-2 Bandwidth 400 kb/s 165 % provisioning 110 % provisioning

8 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Extension to Peer-to-Peer “Sender” peer “Receiver” peer

9 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Multi-Sender Scenario Simulations over ns-2 Bandwidth reserved at each sender 200 kb/s Latency constraint T=0.4 s Simulations over ns-2 Bandwidth reserved at each sender 200 kb/s Latency constraint T=0.4 s ForemanSalesman

10 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st sender 380 kb/s, 36 dB Highest sustainable video quality 420 kb/s, 33.7 dB Foreman Sequence

11 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Distortion model for throughput-limited streaming –Server-client scenario –Extension to peer-to-peer Packet scheduling for low-latency video streaming –Congestion-distortion optimized (CoDiO) scheduling –Low complexity (CoDiO light) scheduler Peer-to-peer video streaming –Distributed peer-to-peer protocol –Receiver-driven CoDiO light retransmission requests –Sender-driven CoDiO light prioritization Presentation Outline

12 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Streaming over a Bottleneck Link Random cross traffic Low bandwidth uplink Video traffic Acknowledgments High bandwidth links C

13 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Related Work TCP-friendly rate-control –Indicates average rate as a function of collected statistics –Does not indicate any particular schedule Rate-Distortion optimized scheduling (RaDiO) –Formalization of the multimedia scheduling problem –Adapted the framework to video streaming –Large gains compared to sequential scheduler [Chou and Miao, 2001] [Chakareski and Girod, ] [Kalman and Girod ] [Floyd et al., 1997]

14 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Congestion-Distortion Optimized Scheduling (CoDiO) Congestion as a new metric –Adaptive to network conditions –Reflects the impact of a sender –Unbounded as it reaches capacity Decide which packets to send (and when) to maximize picture quality while minimizing network congestion C Rate Congestion Δ (seconds) Principle of CoDiO

15 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Estimating Self Congestion defined as average end-to-end delay t0t0 t1t1 t2t2 t3t3 t4t4 t5t5 t6t6 Size of bottleneck queue R(t 3 ) decrease rate C, the capacity R(t 0 ) R(t 1 ) R(t 4 ) average queue size

16 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Decoding buffer state sets the display outcome Expected distortion over a horizon of M frames is Predicting Video Distortion PPIPPP Original picturePicture shownError picture

17 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Parameterized Delay Distribution Probability distribution delay Self congestion affects bottleneck queue length Model distribution by exponential with varying shift

18 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Finding the Best Transmission Order Optimal schedule is Why is this difficult ? –Large search space –Tight coupling between schedules How to solve this problem ?

19 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Find a Good Schedule… at Random! I BB B P B IB B P Pictures to send Schedule IP BB B PI B B BI B

20 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 CoDiO Light Scheduler Select iteratively most important video packet to transmit or retransmit Space transmissions to limit congestion over bottleneck Sufficiently simple to be run at each peer PBPPPBBPPIBBBP… … PBPPPBBPPIBBBP … … PPPPBPPIBBBP… … BBP

21 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 CoDiO Scheduling Performance Simulations over ns-2 Packet loss rate 2% Bandwidth 400 kb/s Propagation delay: 50ms Simulations over ns-2 Packet loss rate 2% Bandwidth 400 kb/s Propagation delay: 50ms 30 % 25 % Mother & DaughterNews

22 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 kb/s Bandwidth 400 kb/s, propagation delay 50 ms 2 % packet loss 0.6 second latency CoDiO ARQ

23 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 CoDiO vs. RaDiO Sequence: Mother & Daughter Packet loss rate 2% Bandwidth 400 kb/s Propagation delay: 50ms Sequence: Mother & Daughter Packet loss rate 2% Bandwidth 400 kb/s Propagation delay: 50ms 40 %

24 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Sequence: Foreman Packet loss rate 2% Bandwidth 400 kb/s Propagation delay: 50ms Sequence: Foreman Packet loss rate 2% Bandwidth 400 kb/s Propagation delay: 50ms 60 % CoDiO vs. RaDiO

25 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Distortion model for throughput-limited streaming –Server-client scenario –Extension to peer-to-peer Packet scheduling for low-latency video streaming –Congestion-distortion optimized (CoDiO) scheduling –Low complexity (CoDiO light) scheduler Peer-to-peer video streaming –Distributed peer-to-peer protocol –Receiver-driven CoDiO light retransmission requests –Sender-driven CoDiO light prioritization Presentation Outline

26 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Join Procedure Initial join Send request to source peer Receive information on session –List of connected peers –Video source rate –Bandwidth provisioning factor –Number of multicast trees Probe peers Find peers with available bandwidth Select parents from candidates to –Maximize diversity –Minimize tree height Connect to multicast trees Request attachment Bandwidth reserved at each parent

27 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Disconnections Detection Reconnection Loss mitigation

28 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Experimental Setup Network/protocol simulation in ns-2 –300 active peers –Random peer arrival/departure average: ON (5 min)/OFF (30 s) –Over-provisioned backbone –Typical access bandwidth distribution –Delay: 5 ms/link + congestion Video streaming –H.264/AVC 250 kb/s –15 minute live multicast [Sripanidkulchai et al., 2004] Downlink UplinkPercentage 512 kb/s256 kb/s 56% 3 Mb/s384 kb/s 21% 1.5 Mb/s896 kb/s 9% 20 Mb/s 2 Mb/s 3% 20 Mb/s 5 Mb/s 11% Downlink UplinkPercentage 512 kb/s256 kb/s 56% 3 Mb/s384 kb/s 21% 1.5 Mb/s896 kb/s 9% 20 Mb/s 2 Mb/s 3% 20 Mb/s 5 Mb/s 11%

29 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Join and Rejoin Latencies Simulations over ns-2, 300 peers Number of trees: 4 Retransmissions enabled Simulations over ns-2, 300 peers Number of trees: 4 Retransmissions enabled

30 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Protocol Scalability Simulations over ns-2 Video traffic Control traffic Protocol overhead Simulations over ns-2 Video traffic Control traffic Protocol overhead

31 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 PPIBBBP … … Receiver-Driven CoDiO Light Retransmission Requests Determine missing packets Iteratively request most important packet Limit number of unacknowledged retransmissions PPIBBBP … …

32 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 % peers connected to all 4 trees With receiver-driven CoDiO light Without retransmissions Performance of Receiver-Driven CoDiO Light

33 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 % peers connected to all 4 trees With content-oblivious retransmissions Without retransmissions Performance of Content-Oblivious Retransmission Scheme

34 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Receiver-driven CoDiO light No retransmissions P2P Video Multicast: 64 out of 300 Peers 250 kb/s 2 second latency for all streams

35 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 P2P Video Multicast: 64 out of 300 Peers 250 kb/s 2 second latency for all streams Receiver-driven CoDiO light No retransmissions

36 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Scheduler iteratively selects Intervals between transmission sufficient to –Mitigate any congestion of the uplink –Reserve rate for control traffic Sender-Driven CoDiO Light Prioritization Sender PI BP B P B

37 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Performance Comparison Simulations over ns-2, 300 peers Number of trees: 4 Retransmissions enabled Simulations over ns-2, 300 peers Number of trees: 4 Retransmissions enabled 30 % 40 % Foreman Mother & Daughter

38 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Simulations over ns-2, 75 peers Number of trees: 4 Retransmissions enabled Simulations over ns-2, 75 peers Number of trees: 4 Retransmissions enabled 35 % 50 % Foreman Mother & Daughter Performance Comparison

39 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Sender-driven CoDiO light dB Without prioritization dB 250 kb/s 0.8 second latency for all streams Average Video Sequence for 75 Peers

40 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Summary of Contributions Distortion model for throughput-limited streaming, captures influence of: –Capacity –Latency constraint –Sequence dependence Packet scheduling for low-latency video streaming –Demonstrated the benefits of considering congestion as a metric –Congestion-distortion optimized packet scheduling Reduces latency vs. ARQ by up to 30% Reduces congestion vs. RaDiO by up to 60% –Evaluated RaDiO and CoDiO in a realistic network simulator –Developed real-time low-complexity CoDiO light scheduler Optimized packet scheduling for peer-to-peer networks –Optimized retransmission scheduling scheme –Prioritization from a sender to its receivers –CoDiO light reduces latency by up to 50%

41 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Main Publications Distortion model for throughput-limited streaming –Setton, Zhu & Girod, Proceedings ICME 2004 –Setton, Zhu & Girod, Proceedings ICIP 2004 –Zhu, Setton & Girod, Proceedings PCS 2004 –Zhu, Setton & Girod, Signal Processing: Image Communications, 2005 Packet scheduling for low-latency video streaming –Setton & Girod, Proceedings MMSP 2004 –Setton, Zhu & Girod, Proceedings ISCAS 2005 –Setton, Yoo, Zhu, Goldsmith & Girod, Wireless Communications Magazine, 2005 Optimized packet scheduling for peer-to-peer networks –Setton, Noh & Girod, Proceedings ACM Multimedia, 2005 –Setton, Noh & Girod, to appear Proceedings ICIP 2006

42 Eric Setton: Congestion-aware video streaming over peer-to-peer networks January 31 st 2006 Other Contributions (not covered in this presentation) Encoding of H.264/AVC SP and SI pictures –Rate-Distortion model for SP and SI pictures –Optimal SP encoder settings for streaming with packet losses –Contributed SP and SI encoder to H.264/AVC standard reference software Adaptive streaming of SP and SI pictures Streaming of SP and SI pictures over peer-to-peer networks Setton & Girod, IEEE Transactions CSVT, accepted 2006 Setton, Ramanathan & Girod, Proceedings VCIP 2006 Setton & Girod, Proceeding VCIP 2005 Setton, Noh & Girod, Proceedings ACM Multimedia, 2005