DYNAMIC ADAPTIVE STREAMING              OVER HTTP @ ITEC                            Christopher Müller, Stefan Lederer and...
AGENDA       (Short) Introduction to DASH & Motivation       Dataset & DASHEncoder       Peer-Assisted DASH       DASH...
MOTIVATION       HTTP Streaming has become very popular on        the Internet               Media encoded a several bit...
DASH ARCHITECTURE   I. Sodagar, “The MPEG-DASH Standard for Multimedia Streaming Over the   Internet”, IEEE Multimedia, IE...
AGENDA       (Short) Introduction to DASH & Motivation       Dataset & DASHEncoder       Peer-Assisted DASH       DASH...
DATASET      Dataset with DASH Content                 Long sequences in high quality                 Various segment-l...
DATASET SEQUENCES                    Name                       Source Quality                  Length         Genre      ...
DASH DATASET                            SEQUENCESChristopher Müller and Stefan Lederer   Dynamic Adaptive Streaming over H...
DASH CONTENT TYPES      Bitrates from 50 kbit/s. to 8 Mbit/s.      Segment Size:              Seconds: 1, 2, 4, 6, 10, ...
DASH ENCODER                            • h.264:      x264 / ffmpeg                            • AAC:        ffmpeg       ...
NON-/PERSISTENT CONN.Christopher Müller and Stefan Lederer   Dynamic Adaptive Streaming over HTTP   11
BENEFITS OF THE DATASET?       Public available and free DASH content       Common basis for evaluations               ...
AGENDA       (Short) Introduction to DASH & Motivation       Dataset & DASHEncoder       Peer-Assisted DASH       DASH...
PEER ASSISTED STREAMING       Clients offer their downloaded segments               Segment requests are monitored by se...
PEER ASSISTED STREAMINGChristopher Müller and Stefan Lederer   Dynamic Adaptive Streaming over HTTP   15
EVALUATION     OMNet++                Simulation framework                INET framework for protocol stack            ...
EVALUATION SETTINGS                                                                     Bitrate        Resolution         ...
SIMULATION RESULTS - SERVER                                        - 15 %                                     - 25 %Christ...
MAJOR FINDINGS       First evaluation simulation:               Up to 25 % bandwidth savings               Directly con...
AGENDA       (Short) Introduction to DASH & Motivation       Dataset & DASHEncoder       Peer-Assisted DASH       DASH...
METHODOLOGY Experiment 1 / Track 1 (601 seconds)       Drive on the freeway A2, passing by the city of Villach in the   ...
EXPERIMENTAL SETUP Bandwidth Shaping Node       Ubuntu 11.04       Linux Hierarchical Token Bucket (htb)       Availab...
MICROSOFT SMOOTH STREAMING Few Switches with a good average bitrate Nevertheless close to unsmoothness at second 300 Chr...
ADOBE DYNAMIC STREAMING High number of unsmooth seconds Rather binary and unpredictableChristopher Mueller   An Evaluati...
APPLE HTTP LIVE STREAMING Very few switches with a lower bitrate Large buffer for energy awareness Christopher Mueller  ...
MPEG – DASH Non stepwise switching Good average bitrate and stable bufferChristopher Mueller   An Evaluation of DASH in ...
COMPARISON     Name              Average Bitrate           Average Switches                    Average Unsmoothness       ...
MAJOR FINDINGS      Microsoft Smooth Streaming: performs best              Altough they don„t use pipelining, maybe spec...
AGENDA       (Short) Introduction to DASH & Motivation       Dataset & DASHEncoder       Peer-Assisted DASH       DASH...
HTTP://DASH.ITEC.AAU.ATChristopher Müller and Stefan Lederer   Dynamic Adaptive Streaming over HTTP   30
STATISTICS                                                      Besucher pro WocheChristopher Müller and Stefan Lederer   ...
DASH @ ITEC       News:               24 Posts about new features, conferences, etc.               ~ 150 comments + a h...
DASH RESEARCH @ ITEC           Publications                  Stefan Lederer, Christopher Müller and Christian Timmerer, ...
THANK YOU FOR YOUR ATTENTION                                   http://dash.itec.aau.atChristopher Müller and Stefan Ledere...
Upcoming SlideShare
Loading in...5
×

ITEC DASH

3,848

Published on

0 Comments
5 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
3,848
On Slideshare
0
From Embeds
0
Number of Embeds
3
Actions
Shares
0
Downloads
37
Comments
0
Likes
5
Embeds 0
No embeds

No notes for slide

Transcript of "ITEC DASH"

  1. 1. DYNAMIC ADAPTIVE STREAMING OVER HTTP @ ITEC Christopher Müller, Stefan Lederer and Christian Timmerer Alpen-Adria Universität Klagenfurt (AAU)  Faculty of Technical Sciences (TEWI) Institute of Information Technology (ITEC)  Multimedia Communication (MMC) 19.03.2012Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 1
  2. 2. AGENDA  (Short) Introduction to DASH & Motivation  Dataset & DASHEncoder  Peer-Assisted DASH  DASH under Vehicular Mobility  DASH @ ITECChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 2
  3. 3. MOTIVATION  HTTP Streaming has become very popular on the Internet  Media encoded a several bitrates, resolutions etc.  Clients request portions of the media due to bandwidth conditions on-demand  Easy to use existing CDN structure  No NAT/Firewall issues due to HTTP  Various technologies  BUT: no standard in use!Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 3
  4. 4. DASH ARCHITECTURE I. Sodagar, “The MPEG-DASH Standard for Multimedia Streaming Over the Internet”, IEEE Multimedia, IEEE MultiMedia, October–December 2011, pp. 62– 67.Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 4
  5. 5. AGENDA  (Short) Introduction to DASH & Motivation  Dataset & DASHEncoder  Peer-Assisted DASH  DASH under Vehicular Mobility  DASH @ ITECChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 5
  6. 6. DATASET  Dataset with DASH Content  Long sequences in high quality  Various segment-length versions  Free available for DASH experiments  PSNR values per frame  Problem: Content Rights  CC-Attribution 2.0 Generic (CC-BY 2.0) License or similar  Free to Share, Free to Remix  Note: YouTube introduces CC-BY in June 2011!  Negotiation with content ownerChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 6
  7. 7. DATASET SEQUENCES Name Source Quality Length Genre Big Buck Bunny 1080p YUV 09:46 Animation Elephants Dream 1080p YUV 10:54 Animation Red Bull Playstreets 1080p, 6 Mbit H.264 01:37:28 Sport The Swiss Account 1080p, 6 Mbit H.264 57:34 Sport Valkaama 1080p, 6 Mbit H.264 01:33:05 Movie Of Forest and Men SD 10:53 MovieChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 7
  8. 8. DASH DATASET SEQUENCESChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 8
  9. 9. DASH CONTENT TYPES  Bitrates from 50 kbit/s. to 8 Mbit/s.  Segment Size:  Seconds: 1, 2, 4, 6, 10, 15  File Organization  Segmented  One file per representation, Byte Range Requests  e.g.: Big Buck Bunny  120 Encodings needed  Only 6 DASH Encoder runsChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 9
  10. 10. DASH ENCODER • h.264: x264 / ffmpeg • AAC: ffmpeg Encode • [VP8/Webm encoding] • MP4Box: Video / Audio / Video + Audio • [Webm segmentation] Container • Generate one MPD • Subfolder Organization • MPD Variation (Byte Range Requests,etc.) MPDChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 10
  11. 11. NON-/PERSISTENT CONN.Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 11
  12. 12. BENEFITS OF THE DATASET?  Public available and free DASH content  Common basis for evaluations  DASH Implementations, Stream Switching Algorithms, Network and Cache Configurations, ...  Enables objective comparison of research results  Also used in the Peer Assisted Streaming evaluation  Provides usefull hints and practices of DASH content generation  Publication: S. Lederer, C. Müller and C. Timmerer, “Dynamic Adaptive Streaming over HTTP Dataset”, In Proceedings of the ACM Multimedia Systems Conference 2012, Chapel Hill, North Carolina, February 22-24, 2012.Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 12
  13. 13. AGENDA  (Short) Introduction to DASH & Motivation  Dataset & DASHEncoder  Peer-Assisted DASH  DASH under Vehicular Mobility  DASH @ ITECChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 13
  14. 14. PEER ASSISTED STREAMING  Clients offer their downloaded segments  Segment requests are monitored by server  Integration in DASH MPD for future clients  Reduction of server load: Goal 10 - 20 %  Peer Traffic  Unsymmetrical network connection  Bottleneck: low upload resourcesChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 14
  15. 15. PEER ASSISTED STREAMINGChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 15
  16. 16. EVALUATION  OMNet++  Simulation framework  INET framework for protocol stack  HTTP Client/Server implementation  DASH Client  MPD Generator + Segment Tracker using external MySQL DatabaseChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 16
  17. 17. EVALUATION SETTINGS Bitrate Resolution 101 kbit/s. 320x240 201 kbit/s. 480x360 395 kbit/s. 480x360 700 kbit/s. 854x480 1172 kbit/s. 853x480 1992 kbit/s. 1280x720 2995 kbit/s. 1920x1080 3992 kbit/s. 1920x1080 4979 kbit/s. 1920x1080 5936 kbit/s. 1920x1080Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 17
  18. 18. SIMULATION RESULTS - SERVER - 15 % - 25 %Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 18
  19. 19. MAJOR FINDINGS  First evaluation simulation:  Up to 25 % bandwidth savings  Directly convertable to CDN cost reduction  Much more possibilities  Intelligent client clustering in larger scale environments  Peer management & download algorithm improvements  MPD update improvementsS. Lederer, C. Müller and C. Timmerer, “Peer-Assisted Dynamic Adaptive Streamingover HTTP - System Design and Evaluation”, Packet Video Workshop 2012 (PV2012), München, Germany, May 10-11, 2012 (to appear).Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 19
  20. 20. AGENDA  (Short) Introduction to DASH & Motivation  Dataset & DASHEncoder  Peer-Assisted DASH  DASH under Vehicular Mobility  DASH @ ITECChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 20
  21. 21. METHODOLOGY Experiment 1 / Track 1 (601 seconds)  Drive on the freeway A2, passing by the city of Villach in the direction to Klagenfurt. Experiment 2 / Track 2 (575 seconds)  From the Alpen-Adria-Universität Klagenfurt on the freeway A2 until the service area around Techelsberg. Experiment 3 / Track 3 (599 seconds)  From the service area around Techelsberg on the freeway A2 to the exit of Klagenfurt. Christopher Mueller An Evaluation of DASH in Vehicular Environments 21
  22. 22. EXPERIMENTAL SETUP Bandwidth Shaping Node  Ubuntu 11.04  Linux Hierarchical Token Bucket (htb)  Available bandwidth will be adjusted every 2 seconds Network Emulation Node  Emulates a round trip time of 150ms Server  Server based on Windows Server 2008 and IIS Client  Windows or Linux depending on the evaluation system Christopher Mueller An Evaluation of DASH in Vehicular Environments 22
  23. 23. MICROSOFT SMOOTH STREAMING Few Switches with a good average bitrate Nevertheless close to unsmoothness at second 300 Christopher Mueller An Evaluation of DASH in Vehicular Environments 23
  24. 24. ADOBE DYNAMIC STREAMING High number of unsmooth seconds Rather binary and unpredictableChristopher Mueller An Evaluation of DASH in Vehicular Environments 24
  25. 25. APPLE HTTP LIVE STREAMING Very few switches with a lower bitrate Large buffer for energy awareness Christopher Mueller An Evaluation of DASH in Vehicular Environments 25
  26. 26. MPEG – DASH Non stepwise switching Good average bitrate and stable bufferChristopher Mueller An Evaluation of DASH in Vehicular Environments 26
  27. 27. COMPARISON Name Average Bitrate Average Switches Average Unsmoothness [kbps] [Number of Switches] [Seconds] Microsoft 1522 51 0 Adobe 1239 97 64 Apple 1162 7 0MPEG – DASH 1464 166 0 Pipelined Christopher Mueller An Evaluation of DASH in Vehicular Environments 27
  28. 28. MAJOR FINDINGS  Microsoft Smooth Streaming: performs best  Altough they don„t use pipelining, maybe specific TCP implementation?  Adobe Dynamic Streaming: not usable, interesting buffer  Apple HTTP Streaming: interesting features  MPEG-2 TS, Large Buffer, Conservative Approach (Energy?)  MPEG-DASH: good start with rather simple algorithm  Pipelining brings improvement + integrate further features: parallel downloads, TCP modification, other Protocols (e.g. SPDY) C. Müller, S. Lederer and C. Timmerer, “An Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular Environments”, ACM Workshop on Mobile Video, Chapel Hill, North Carolina, February 24, 2012.Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 28
  29. 29. AGENDA  (Short) Introduction to DASH & Motivation  Dataset & DASHEncoder  Peer-Assisted DASH  DASH under Vehicular Mobility  DASH @ ITECChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 29
  30. 30. HTTP://DASH.ITEC.AAU.ATChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 30
  31. 31. STATISTICS Besucher pro WocheChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 31
  32. 32. DASH @ ITEC  News:  24 Posts about new features, conferences, etc.  ~ 150 comments + a huge number of emails  Cooperations  VideoLAN VLC  Mozilla  DASH-PG  Tools:  DASH VLC Plugin  DASH Dataset  libDASH  DASH MPD Validator (Markus  DASHEncoder Waltl)Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 32
  33. 33. DASH RESEARCH @ ITEC  Publications  Stefan Lederer, Christopher Müller and Christian Timmerer, “Peer-Assisted Dynamic Adaptive Streaming over HTTP – System Design and Evaluation“, In Proceedings of the IEEE International Packet Video Workshop 2012, Munich, Germany, May 10-11, 2012. (to appear)  Christopher Müller, Stefan Lederer and Christian Timmerer, “An Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular Environments”, In Proceedings of the ACM Multimedia Systems Conference 2012 and the 4th ACM Workshop on Mobile Video, Chapel Hill, North Carolina, February 24, 2012.  Stefan Lederer, Christopher Müller and Christian Timmerer, “Dynamic Adaptive Streaming over HTTP Dataset”, In Proceedings of the ACM Multimedia Systems Conference 2012, Chapel Hill, North Carolina, February 22-24, 2012.  Christopher Müller and Christian Timmerer, “A VLC Media Player Plugin enabling Dynamic Adaptive Streaming over HTTP”, In Proceedings of the ACM Multimedia 2011 , Scottsdale, Arizona, November 28, 2011.  Christopher Müller and Christian Timmerer, “A Test-Bed for the Dynamic Adaptive Streaming over HTTP featuring Session Mobility”, In Proceedings of the ACM Multimedia Systems Conference 2011, San Jose, California, February 23-25, 2011.  Christian Timmerer and Christopher Müller, “HTTP Streaming of MPEG Media”, In Proceedings of the Streaming Day 2010, Udine, Italy, September 16-17, 2010.  Patents  Christopher Müller, Yuwen He, James Crenshaw, Bandwidth Adaptation for Dynamic Adaptive Transfering of Multimedia, U.S. Provisional Application No.: 61/576,334.Christopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 33
  34. 34. THANK YOU FOR YOUR ATTENTION http://dash.itec.aau.atChristopher Müller and Stefan Lederer Dynamic Adaptive Streaming over HTTP 34

×