Efficient Path Aggregation and Error Control for Video

Efficient Path Aggregation and Error Control for Video Streaming OMESH TICKOO, Shiv Kalyanaraman, John Woods Rensselaer Polytechnic Institute (RPI) : “shiv rpi” Rensselaer Polytechnic Institute 1 Shivkumar Kalyanaraman Sponsors: ARO, DARPA-NMS, Intel

Introduction q q q Motivation: Video over best-effort Internet q Broadband => more access bandwidth q End-to-end (E 2 E) => constraints due to path congestion q Virtual extension of broadband access pipe E 2 E using multi-paths Path Diversity: dimensions q Aggregate Capacity q Delay diversity q Loss diversity q Correlations in path performance characteristics Key: Match inherent content diversity to path diversity Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 2

Motivation: Internet Path Congestion limits E 2 E bandwidth Internet Performance Server Access Link Client Access Link Performance Saturation (even w/ many flows/path) Shivkumar Kalyanaraman Rensselaer Polytechnic Institute Access 3 Link Speed

Multi-paths? Overlays or peers can provide path diversity even if multi-paths not available natively in the Internet. Issue: diversity of performance (b/w, delay, loss), possible correlations… Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 4

Smart Multi-path Capacity Aggregation (SMCA): Motivation Path (Flow) Aggregator/ Multiplexer Performance Server Access Link Rensselaer Polytechnic Institute Internet Path (Flow) Aggregator/ Demultiplexer E 2 E Broadband Virtual Pipe Abstraction!! Client Access Link Performance Scaling Access 5 Link Speed Shivkumar Kalyanaraman

Single path issues: capacity, delay, loss… High Delay/Jitter Low Capacity Lossy Network paths usually have: • low e 2 e capacity, • high latencies and • high/variable loss rates. Rensselaer Polytechnic Institute 6 Time Shivkumar Kalyanaraman

SMCA: Leverage Diversity! Low Perceived Loss High Perceived Capacity Low Perceived Delay/Jitter Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 7

SMCA: Framework Content Delay Diversity Unit Loss Diversity Unit Network Receive Buffer Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 8

SMCA: Delay Diversity Unit High Delay RANK Low Delay RANK Application Data Paths Ranked by Latency Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 9

SMCA: Delay Diversity Unit High Delay RANK Low Delay RANK Application Data Paths Ranked by Latency Early deadline packets mapped to low-delay paths Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 10

SMCA: Delay Diversity Unit High Delay RANK Transmit Queue Paths Ranked by Latency Low Delay RANK Early deadline packets (in order of rank) mapped to low-delay paths (in order. Shivkumar of rank)Kalyanaraman Rensselaer Polytechnic Institute 11

SMCA: Delay Diversity Unit High Delay RANK Low Delay RANK Transmit Queue Paths Ranked by Latency Late deadline packets mapped to high-delay paths… Note: these packets leave the sender roughly at the. Kalyanaraman same Shivkumar time as the early-deadline packets 12 Rensselaer Polytechnic Institute

SMCA: Delay Diversity Loss Diversity High Delay Low Delay Transmit Queue Paths Ranked by Latency Consider a delay-based group of paths and the associated packets… Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 13

SMCA: Delay Diversity Loss Diversity High Delay Low Delay Transmit Queue Paths Ranked by Latency Consider a delay-based group of paths and the associated packets… Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 14

SMCA: Loss Diversity Unit High Loss RANK n GOPs Low Loss RANK Paths Ranked by Loss Rate Re-rank Paths within this group based upon packet loss rates Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 15

SMCA: Loss Diversity Unit P B B I High Loss RANK Low Loss RANK n GOPs Paths Ranked by Loss Rate Enlarged View of Packets (with content labels) and Paths Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 16

SMCA: Loss Diversity Unit P B B I High Loss RANK Low Loss RANK n GOPs Paths Ranked by Loss Rate Map high priority packets (eg: I-frame packets) to low loss rate rank paths Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 17

SMCA: Loss Diversity Unit P B B I High Loss RANK Low Loss RANK n GOPs Paths Ranked by Loss Rate Continue map packets to low loss rank paths based upon priority (Eg: P-frames get the next set of loss-ranked Shivkumarpaths) Kalyanaraman Rensselaer Polytechnic Institute 18

SMCA: Loss Diversity Unit P B B I High Loss RANK Low Loss RANK n GOPs Paths Ranked by Loss Rate Lowest priority packets get high loss rate paths (within the delay-based group of paths) Kalyanaraman Shivkumar Rensselaer Polytechnic Institute 19

SMCA: Loss Diversity Unit + FEC P-FEC I-FEC P B B I High Loss RANK Low Loss RANK n GOPs Paths Ranked by Loss Rate FEC (unequal FEC) for a GOP mapped within the same delay-group, but mapped to the higher loss paths Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 20

SMCA: Performance with increasing number of Paths Content Source Content Sink Background traffic generator Background traffic sink Num. Of Paths PSNR (d. B) 1 2 3 4 5 20. 98 22. 48 25. 42 26. 02 28. 04 Table 1. Average PSNR Variation with Number of Paths Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 21

Topology to test delay diversity and loss diversity gains Content Source Content Sink 5 paths Background traffic generator Background traffic sink Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 22

SMCA gains with delay diversity Avg. Delay (ms) 300 SMCA PSNR(d. B) PT PSNR(d. B) OPMS PSNR(d. B) 21. 78 18. 73 11. 03 100 25. 12 24. 21 19. 19 50 28. 32 29. 46 24. 33 30 30. 12 31. 63 27. 96 Table 3. Gains with Delay Variation SMCA achieves even better performance (than simple multipath mapping: OPMS) when average delay and jitter is higher Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 23

SMCA gains with loss diversity Avg. Loss Prob. SMCA PSNR(d. B) PT PSNR(d. B) OPMS PSNR(d. B) 0. 4 22. 78 20. 31 11. 64 0. 35 26. 32 26. 86 18. 21 0. 1 29. 03 29. 02 24. 43 0. 05 29. 32 31. 82 26. 06 Table 2. Gains with Loss Variation SMCA achieves even better performance (than simple multipath mapping: OPMS) when average loss and loss variations are higher! Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 24

Summary q Multi-path performance diversity can be leveraged E 2 E q q q Ideas: q q q Key: must be mapped to content diversity (Similar to lessons learnt from content-driven unequal FEC protection vs uniform FEC protection) Map late deadline packets to high latency paths Map higher priority packets to lower loss rate paths (within a delaybased group of paths) q FEC packets sent on paths different from that of associated content (FEC: lower priority) Our scheme can scale to handle lots of paths q q Possible with p 2 p networks (eg: 10 -100 kbps from single path, but 10 s of paths) Does not require MD coding, or high complexity optimization Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 25