Proxy Cache Management for Fine-Grained Scalable Video Streaming - PowerPoint PPT Presentation

andrew
proxy cache management for fine grained scalable video streaming l.
Skip this Video
Loading SlideShow in 5 Seconds..
Proxy Cache Management for Fine-Grained Scalable Video Streaming PowerPoint Presentation
Download Presentation
Proxy Cache Management for Fine-Grained Scalable Video Streaming

play fullscreen
1 / 28
Download Presentation
Proxy Cache Management for Fine-Grained Scalable Video Streaming
359 Views
Download Presentation

Proxy Cache Management for Fine-Grained Scalable Video Streaming

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

  1. Proxy Cache Management for Fine-Grained Scalable Video Streaming Jiangchuan Liu The Chinese University of Hong Kong Xiaowen Chu and Jianliang Xu Baptist University of Hong Kong Infocom’04, March 2003

  2. Outline • Introduction and Motivations • Problem Settings and Solutions • Performance Evaluation and Comparison • Conclusion and Future Work

  3. Video Caching • Proxy caching saving video objects at proxies close to clients • Temporal locality • Geographical locality

  4. Unique Features Video objects vs. Web objects • High data rate, yet adaptive • Long playback duration ► Various interactions: • random access • early termination ► Huge volume • one-hour MPEG-1, about 675 MB

  5. r2 r1 r2 r1 2 3 1 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Partial Caching • Interval caching (Dan96, Tewari98…) • Prefix caching (Sen99, Jin02…) • Segment caching (Wu01, Miao02, Chae,Chen03…)

  6. Common Assumptions • Continuous playback • No interactions or premature termination • Homogeneous segment access rate • Homogeneous clients • Identical access bandwidth • Time partitioning only • Non-adaptive caching • Non-scalable single-rate video

  7. Our Assumptions and Objectives • Assumptions • User interactivity: early terminations • Heterogeneous client access bandwidths • Vehicle • Fine-Grained Scalable (FGS) video • Objectives • Rate adaptive caching and streaming • Understanding the benefits (?) of FGS caching • Vs. Replication (Hartanto02), Transcoding (Tang02)

  8. Related Work and Differences cut-off rate • Video staging (Zhang00) • Quality adaptation (Yu00, Rejaie00) • Adaptation per user basis • Replacement • Prefetching • Blocking performance (Kangasharju02) • Admission control • Homogeneous access rate • Caching for VCR-operations (Fahmi01)

  9. Outline • Introduction and Motivations • Problem Settings and Solutions • Performance Evaluation and Comparison • Conclusion and Future Work

  10. System Model and Operations

  11. Model Parameters

  12. Problem(1): Caching Strategy • Explore the rate adaptability of FGS in caching • Problem: • Given cache size and client utility level, which portion of an FGS video should be cached • Objective • Min transmission cost • Difficulty: • Heterogeneous bandwidth demands • Non-uniform segment access rate • A 2-D space: time and rate • Greedy is not optimal

  13. Problem(1): Solution • 2-segment • Exhaustive search • Multi-segment • Access probability – segment/rate

  14. Problem (2): Utility Assignment • Explore the rate adaptability of FGS in both caching and streaming • Problem • Given cache size and backbone bandwidth limits, jointly decide the caching strategy and utility assignment of each client. • Objective • Max expected client utility , , , ,

  15. Problem(2): Iterative Solution • Difficulty • Utility assignment  optimal caching strategy (problem 1) • Caching strategy  optimal utility assignment (dynamic programming) • Iterative optimization • 2-Segment – exhaustive search on cache partition

  16. Optimization for Multiple Objects • Heterogeneity of objects • Access rate • Access bandwidth • Client distribution • Cache partitioning • Backbone Bandwidth partitioning • 2-D Knapsack • Pseudo-polynomial partitioning algorithm

  17. Outline • Introduction and Motivations • Problem Settings and Solutions • Performance Evaluation and Comparison • Conclusion and Future Work

  18. Sample Configuration • Client • Multiple classes • Uniform, skewed • Utility function • Linear • 2-segments: early termination • Probability = 0.3 • A conservative configuration !

  19. Backbone Bandwidth Reduction • MaxLen: length first • MaxRate: rate first

  20. Joint Optimal Caching & Utility Assignment

  21. Results for Multiple Videos • Baseline • uniform cache partition + proportional bandwidth partition

  22. Scalable Video or Replicated Video? • Optimal caching for replicated video • 1D knapsack • Backbone bandwidth reduction with FGS

  23. Scalable Video or Transcoding ? Given a frame interval of 30 ms, our PC can support about 300 concurrent filter/assembler operations

  24. Outline • Introduction and Motivations • System Description • Problem Settings and Solutions • Performance Evaluation and Comparison • Conclusion and Future Work

  25. Conclusion • FGS-based proxy caching • Key problems • Optimal caching strategy • Optimal utility assignment • Optimization for multiple videos • Performance Evaluation • Backbone bandwidth reduction • Utility improvement • Comparision • FGS caching vs. Replication caching • FGS filtering vs. Transcoding

  26. Future Work • Utility functions • Tradeoff: accuracy/speed • Multi-segments • Fastforward, backward • Practical issues • Error control • Synchronization • Signaling

  27. Thanks Q & A

  28. Scalable Video or Replicated Video? • Utility improvementwith FGS