SlideShare une entreprise Scribd logo
1  sur  30
Télécharger pour lire hors ligne
DASH in Twitch: 
Adaptive Bitrate 
Streaming in Live Game 
Streaming Platforms 
Karine Pires, Gwendal Simon 
Telecom Bretagne, France 
Nov 2014
Live Streaming : A New Context 
Cloud Delivery Process 
2 / 19 K. Pires, G. Simon DASH in Twitch
Live Streaming : A New Context 
Cloud Delivery Process 
more and more 
non-professional 
broadcasters 
2 / 19 K. Pires, G. Simon DASH in Twitch
Live Streaming : A New Context 
Cloud Delivery Process 
more and more CDN 
heterogeneous 
end-viewers 
2 / 19 K. Pires, G. Simon DASH in Twitch
Live Streaming : A New Context 
Cloud Delivery Process 
CDN 
Data-Center 
2 / 19 K. Pires, G. Simon DASH in Twitch
Live Streaming : A New Context 
Cloud Delivery Process 
CDN 
Data-Center 
Adaptive Streaming 
1 video stream = k representations 
2 / 19 K. Pires, G. Simon DASH in Twitch
Live Streaming : A New Context 
Cloud Delivery Process 
CDN 
Data-Center 
4th in US internet traffic 
+55 millions unique visitors 
2 / 19 K. Pires, G. Simon DASH in Twitch
Two options for the delivery process 
Traditional Process 
Raw 
Video 
Ingest 
Servers 
Delivery 
Servers 
3 / 19 K. Pires, G. Simon DASH in Twitch
Two options for the delivery process 
Traditional Process 
Raw 
Video 
Ingest 
Servers 
Delivery 
Servers 
Transcoding Process 
Raw 
Ingest 
Video 
Servers 
Transcode 
Servers 
Delivery 
Servers 
3 / 19 K. Pires, G. Simon DASH in Twitch
Which live streams to transcode 
Traditional 
Process 
# CPU 
# QoE 
" BW 
Transcoding 
Process 
" CPU 
" QoE 
# BW 
Live stream ? 
Trade-off 
4 / 19 K. Pires, G. Simon DASH in Twitch
Our contributions 
Adaptive Bit-Rate streaming in Twitch : 
Real traces from Twitch 
Decision of which channels should be delivered with ABR 
Comparison of different decision strategies 
5 / 19 K. Pires, G. Simon DASH in Twitch
A Tour on Twitch 
6 / 19 K. Pires, G. Simon DASH in Twitch
What is a broadcaster 
A broadcaster streaming two gaming sessions 
nb. of viewers 
online online 
time 
t1 t01 
t2 t02 
session 1 session 2 
7 / 19 K. Pires, G. Simon DASH in Twitch
Data retrieval 
For 3 months ! from 6 Jan. to 6 Apr. 
time 
all 
channels 
of one 
snapshot 
9 :00 9 :05 9 :10 9 :15 9 :20 
every five minutes ! one snapshot 
Dataset available : http ://dash.ipv6.enstb.fr/dataset/twitch/ 
8 / 19 K. Pires, G. Simon DASH in Twitch
How many online broadcasters 
0 10 20 30 40 50 60 70 80 90 
10K 
8K 
6K 
4K 
2K 
0 
Days 
Nb. of online channels 
min max 
9 / 19 K. Pires, G. Simon DASH in Twitch
How much bandwidth is needed 
0 10 20 30 40 50 60 70 80 90 
2 
1 
0 
Days 
Bandwidth (Tbps) 
min max 
10 / 19 K. Pires, G. Simon DASH in Twitch
Raw videos properties 
240p 360p 480p 720p 1080p 
0.6 
0.4 
0.2 
0 
Representation 
Sessions ratio 
11 / 19 K. Pires, G. Simon DASH in Twitch
Raw videos properties 
240p 360p 480p 720p 1080p 
0.6 
0.4 
0.2 
0 
Representation 
Sessions ratio 
1 
0.8 
0.6 
0.4 
0.2 
0 
0.1 1 2 10 
Video bit-rate (Mbps) 
CDF of the sessions 
480p 
11 / 19 K. Pires, G. Simon DASH in Twitch
Raw videos properties 
240p 360p 480p 720p 1080p 
0.6 
0.4 
0.2 
0 
Representation 
Sessions ratio 
1 
0.8 
0.6 
0.4 
0.2 
0 
0.1 1 2 10 
Video bit-rate (Mbps) 
CDF of the sessions 
480p 720p 1080p 
11 / 19 K. Pires, G. Simon DASH in Twitch
Selecting the ABR 
streams 
12 / 19 K. Pires, G. Simon DASH in Twitch
When to decide a stream should be transcoded 
At-startup 
Strategy 
transcoding decision 
time 
t1 t01 
On-the-fly 
Strategy 
transcoding decision 
time 
t1 t01 
transcoding traditional 
13 / 19 K. Pires, G. Simon DASH in Twitch
Strategies 
None : no ABR 
All : all streams are streamed with ABR 
14 / 19 K. Pires, G. Simon DASH in Twitch
Strategies 
None : no ABR 
All : all streams are streamed with ABR 
Top-50 (at-startup) : streams that were popular 
14 / 19 K. Pires, G. Simon DASH in Twitch
Strategies 
None : no ABR 
All : all streams are streamed with ABR 
Top-50 (at-startup) : streams that were popular 
Threshold-1000 (on-the-fly) immediate popularity 
14 / 19 K. Pires, G. Simon DASH in Twitch
Evaluation settings 
Sessions ! Twitch dataset 
Viewers ! DASH dataset 
ABR ! proportional to raw video 
15 / 19 K. Pires, G. Simon DASH in Twitch
Results : degraded viewers served 
0 10 20 
1 
0.5 
0 
All 
None 
Days 
Viewers ratio 
Top Threshold 
16 / 19 K. Pires, G. Simon DASH in Twitch
Results : costs evaluation 
None All Top Threshold 
150 
100 
50 
0 
-39% -40% 
+5% 
Strategies 
Total Price ($M) 
Delivery Transcode 
17 / 19 K. Pires, G. Simon DASH in Twitch
Conclusion 
18 / 19 K. Pires, G. Simon DASH in Twitch
Discussions 
preliminary work=global picture!details neglected 
more formal and accurate formulation 
more precise QoE estimation 
better transcoding needs model 
management of different hardware resources 
19 / 19 K. Pires, G. Simon DASH in Twitch
Discussions 
preliminary work=global picture!details neglected 
more formal and accurate formulation 
more precise QoE estimation 
better transcoding needs model 
management of different hardware resources 
more efficient strategies can be designed 
sophisticated strategies 
statistical approaches for popularity forecasting 
19 / 19 K. Pires, G. Simon DASH in Twitch

Contenu connexe

En vedette

En vedette (10)

Video production in the cloud for live streaming
Video production in the cloud for live streamingVideo production in the cloud for live streaming
Video production in the cloud for live streaming
 
Live Streaming is the Future and the Future is Now
Live Streaming is the Future and the Future is NowLive Streaming is the Future and the Future is Now
Live Streaming is the Future and the Future is Now
 
Live Streaming: Consumption, branding and it's future
Live Streaming: Consumption, branding and it's futureLive Streaming: Consumption, branding and it's future
Live Streaming: Consumption, branding and it's future
 
How to Build Your Brand with Live Streaming Video
How to Build Your Brand with Live Streaming VideoHow to Build Your Brand with Live Streaming Video
How to Build Your Brand with Live Streaming Video
 
Peer-to-Peer and live streaming: 10 years of academic research
Peer-to-Peer and live streaming: 10 years of academic researchPeer-to-Peer and live streaming: 10 years of academic research
Peer-to-Peer and live streaming: 10 years of academic research
 
Ultimate Guide to Live Streaming
Ultimate Guide to Live StreamingUltimate Guide to Live Streaming
Ultimate Guide to Live Streaming
 
Live Streaming
Live StreamingLive Streaming
Live Streaming
 
Joel Comm - The Live Streaming Revolution
Joel Comm - The Live Streaming RevolutionJoel Comm - The Live Streaming Revolution
Joel Comm - The Live Streaming Revolution
 
Virtual reality and 360° live streaming
Virtual reality and 360° live streamingVirtual reality and 360° live streaming
Virtual reality and 360° live streaming
 
Four Steps to Sure-Fire Live Streaming Success
Four Steps to Sure-Fire Live Streaming SuccessFour Steps to Sure-Fire Live Streaming Success
Four Steps to Sure-Fire Live Streaming Success
 

Similaire à DASH in Twitch: Adaptive Bitrate Streaming in Live Game Streaming Platforms

Streaming Media over the Internet
Streaming Media over the InternetStreaming Media over the Internet
Streaming Media over the Internet
Videoguy
 
Online Video Delivery
Online Video DeliveryOnline Video Delivery
Online Video Delivery
Dan O'Hara
 
Principles of vid tech UNFINISHED
Principles of vid tech UNFINISHEDPrinciples of vid tech UNFINISHED
Principles of vid tech UNFINISHED
Kieran Ryan
 

Similaire à DASH in Twitch: Adaptive Bitrate Streaming in Live Game Streaming Platforms (20)

Adaptive Delivery of Live Video Stream: Infrastructure cost vs. QoE
Adaptive Delivery of Live Video Stream: Infrastructure cost vs. QoEAdaptive Delivery of Live Video Stream: Infrastructure cost vs. QoE
Adaptive Delivery of Live Video Stream: Infrastructure cost vs. QoE
 
Velocity EU 2014 - How BBC Sport Scales Engineering
Velocity EU 2014 - How BBC Sport Scales EngineeringVelocity EU 2014 - How BBC Sport Scales Engineering
Velocity EU 2014 - How BBC Sport Scales Engineering
 
Beginners Guide to Church Streaming .PPTX
Beginners Guide to Church Streaming .PPTXBeginners Guide to Church Streaming .PPTX
Beginners Guide to Church Streaming .PPTX
 
Brightcove live tech overview
Brightcove live tech overviewBrightcove live tech overview
Brightcove live tech overview
 
Portogdg video
Portogdg videoPortogdg video
Portogdg video
 
Streaming Media over the Internet
Streaming Media over the InternetStreaming Media over the Internet
Streaming Media over the Internet
 
Corkgdg video
Corkgdg videoCorkgdg video
Corkgdg video
 
Full HD LINE free call
Full HD LINE free callFull HD LINE free call
Full HD LINE free call
 
Armadajs video
Armadajs videoArmadajs video
Armadajs video
 
2017 LTS catalogue
2017 LTS catalogue2017 LTS catalogue
2017 LTS catalogue
 
Frontcon video
Frontcon videoFrontcon video
Frontcon video
 
Broadcasting the World's Largest Sporting Events: AWS Media Services When It ...
Broadcasting the World's Largest Sporting Events: AWS Media Services When It ...Broadcasting the World's Largest Sporting Events: AWS Media Services When It ...
Broadcasting the World's Largest Sporting Events: AWS Media Services When It ...
 
Online Video Delivery
Online Video DeliveryOnline Video Delivery
Online Video Delivery
 
Following the Sports Fan with Arbitron’s PPM
Following the Sports Fan with Arbitron’s PPMFollowing the Sports Fan with Arbitron’s PPM
Following the Sports Fan with Arbitron’s PPM
 
Principles of vid tech UNFINISHED
Principles of vid tech UNFINISHEDPrinciples of vid tech UNFINISHED
Principles of vid tech UNFINISHED
 
IPTV System Design and Deployment-Updated
IPTV System Design and Deployment-UpdatedIPTV System Design and Deployment-Updated
IPTV System Design and Deployment-Updated
 
Designing and Deploying IPTV Systems
Designing and Deploying IPTV SystemsDesigning and Deploying IPTV Systems
Designing and Deploying IPTV Systems
 
Krakow video
Krakow videoKrakow video
Krakow video
 
Why Game Developers Care About Twitch - TwitchCon Developer Day 2017
Why Game Developers Care About Twitch - TwitchCon Developer Day 2017Why Game Developers Care About Twitch - TwitchCon Developer Day 2017
Why Game Developers Care About Twitch - TwitchCon Developer Day 2017
 
-D
-D-D
-D
 

Plus de Gwendal Simon

Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...
Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...
Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...
Gwendal Simon
 

Plus de Gwendal Simon (13)

Reproducible research at ACM MMSys
Reproducible research at ACM MMSysReproducible research at ACM MMSys
Reproducible research at ACM MMSys
 
Netgames: history and preparing 2018 edition
Netgames: history and preparing 2018 editionNetgames: history and preparing 2018 edition
Netgames: history and preparing 2018 edition
 
Virtual Reality in 5G Networks
Virtual Reality in 5G NetworksVirtual Reality in 5G Networks
Virtual Reality in 5G Networks
 
Research on cloud gaming: status and perspectives
Research on cloud gaming: status and perspectivesResearch on cloud gaming: status and perspectives
Research on cloud gaming: status and perspectives
 
Fast Near-Optimal Delivery of Live Streams in CDN
Fast Near-Optimal Delivery of Live Streams in CDNFast Near-Optimal Delivery of Live Streams in CDN
Fast Near-Optimal Delivery of Live Streams in CDN
 
Scadoosh: Scaling Down the Footprint of Rate-Adaptive Live Streaming on CDN I...
Scadoosh: Scaling Down the Footprint of Rate-Adaptive Live Streaming on CDN I...Scadoosh: Scaling Down the Footprint of Rate-Adaptive Live Streaming on CDN I...
Scadoosh: Scaling Down the Footprint of Rate-Adaptive Live Streaming on CDN I...
 
Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...
Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...
Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...
 
Time-Shifted TV in Content Centric Networks: the Case for Cooperative In-Netw...
Time-Shifted TV in Content Centric Networks: the Case for Cooperative In-Netw...Time-Shifted TV in Content Centric Networks: the Case for Cooperative In-Netw...
Time-Shifted TV in Content Centric Networks: the Case for Cooperative In-Netw...
 
Internet : pourquoi ça marche
Internet : pourquoi ça marcheInternet : pourquoi ça marche
Internet : pourquoi ça marche
 
Optimal Network Locality in Distributed Services
Optimal Network Locality in Distributed ServicesOptimal Network Locality in Distributed Services
Optimal Network Locality in Distributed Services
 
Cloud Engineering
Cloud EngineeringCloud Engineering
Cloud Engineering
 
peer-to-peer oppotunities
peer-to-peer oppotunitiespeer-to-peer oppotunities
peer-to-peer oppotunities
 
Infrastructureless Wireless networks
Infrastructureless Wireless networksInfrastructureless Wireless networks
Infrastructureless Wireless networks
 

Dernier

Call Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
9953056974 Low Rate Call Girls In Saket, Delhi NCR
 
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
dollysharma2066
 
Integrated Test Rig For HTFE-25 - Neometrix
Integrated Test Rig For HTFE-25 - NeometrixIntegrated Test Rig For HTFE-25 - Neometrix
Integrated Test Rig For HTFE-25 - Neometrix
Neometrix_Engineering_Pvt_Ltd
 
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night StandCall Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
amitlee9823
 
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
dharasingh5698
 
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak HamilCara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Kandungan 087776558899
 

Dernier (20)

Thermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.pptThermal Engineering -unit - III & IV.ppt
Thermal Engineering -unit - III & IV.ppt
 
Unit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdfUnit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdf
 
Call Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Netaji Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
 
A Study of Urban Area Plan for Pabna Municipality
A Study of Urban Area Plan for Pabna MunicipalityA Study of Urban Area Plan for Pabna Municipality
A Study of Urban Area Plan for Pabna Municipality
 
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
 
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
 
Hazard Identification (HAZID) vs. Hazard and Operability (HAZOP): A Comparati...
Hazard Identification (HAZID) vs. Hazard and Operability (HAZOP): A Comparati...Hazard Identification (HAZID) vs. Hazard and Operability (HAZOP): A Comparati...
Hazard Identification (HAZID) vs. Hazard and Operability (HAZOP): A Comparati...
 
Work-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptxWork-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptx
 
University management System project report..pdf
University management System project report..pdfUniversity management System project report..pdf
University management System project report..pdf
 
Integrated Test Rig For HTFE-25 - Neometrix
Integrated Test Rig For HTFE-25 - NeometrixIntegrated Test Rig For HTFE-25 - Neometrix
Integrated Test Rig For HTFE-25 - Neometrix
 
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night StandCall Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
 
22-prompt engineering noted slide shown.pdf
22-prompt engineering noted slide shown.pdf22-prompt engineering noted slide shown.pdf
22-prompt engineering noted slide shown.pdf
 
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
 
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak HamilCara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
 
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
 
Block diagram reduction techniques in control systems.ppt
Block diagram reduction techniques in control systems.pptBlock diagram reduction techniques in control systems.ppt
Block diagram reduction techniques in control systems.ppt
 
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
 
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
 
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdfONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
 
Thermal Engineering Unit - I & II . ppt
Thermal Engineering  Unit - I & II . pptThermal Engineering  Unit - I & II . ppt
Thermal Engineering Unit - I & II . ppt
 

DASH in Twitch: Adaptive Bitrate Streaming in Live Game Streaming Platforms

  • 1. DASH in Twitch: Adaptive Bitrate Streaming in Live Game Streaming Platforms Karine Pires, Gwendal Simon Telecom Bretagne, France Nov 2014
  • 2. Live Streaming : A New Context Cloud Delivery Process 2 / 19 K. Pires, G. Simon DASH in Twitch
  • 3. Live Streaming : A New Context Cloud Delivery Process more and more non-professional broadcasters 2 / 19 K. Pires, G. Simon DASH in Twitch
  • 4. Live Streaming : A New Context Cloud Delivery Process more and more CDN heterogeneous end-viewers 2 / 19 K. Pires, G. Simon DASH in Twitch
  • 5. Live Streaming : A New Context Cloud Delivery Process CDN Data-Center 2 / 19 K. Pires, G. Simon DASH in Twitch
  • 6. Live Streaming : A New Context Cloud Delivery Process CDN Data-Center Adaptive Streaming 1 video stream = k representations 2 / 19 K. Pires, G. Simon DASH in Twitch
  • 7. Live Streaming : A New Context Cloud Delivery Process CDN Data-Center 4th in US internet traffic +55 millions unique visitors 2 / 19 K. Pires, G. Simon DASH in Twitch
  • 8. Two options for the delivery process Traditional Process Raw Video Ingest Servers Delivery Servers 3 / 19 K. Pires, G. Simon DASH in Twitch
  • 9. Two options for the delivery process Traditional Process Raw Video Ingest Servers Delivery Servers Transcoding Process Raw Ingest Video Servers Transcode Servers Delivery Servers 3 / 19 K. Pires, G. Simon DASH in Twitch
  • 10. Which live streams to transcode Traditional Process # CPU # QoE " BW Transcoding Process " CPU " QoE # BW Live stream ? Trade-off 4 / 19 K. Pires, G. Simon DASH in Twitch
  • 11. Our contributions Adaptive Bit-Rate streaming in Twitch : Real traces from Twitch Decision of which channels should be delivered with ABR Comparison of different decision strategies 5 / 19 K. Pires, G. Simon DASH in Twitch
  • 12. A Tour on Twitch 6 / 19 K. Pires, G. Simon DASH in Twitch
  • 13. What is a broadcaster A broadcaster streaming two gaming sessions nb. of viewers online online time t1 t01 t2 t02 session 1 session 2 7 / 19 K. Pires, G. Simon DASH in Twitch
  • 14. Data retrieval For 3 months ! from 6 Jan. to 6 Apr. time all channels of one snapshot 9 :00 9 :05 9 :10 9 :15 9 :20 every five minutes ! one snapshot Dataset available : http ://dash.ipv6.enstb.fr/dataset/twitch/ 8 / 19 K. Pires, G. Simon DASH in Twitch
  • 15. How many online broadcasters 0 10 20 30 40 50 60 70 80 90 10K 8K 6K 4K 2K 0 Days Nb. of online channels min max 9 / 19 K. Pires, G. Simon DASH in Twitch
  • 16. How much bandwidth is needed 0 10 20 30 40 50 60 70 80 90 2 1 0 Days Bandwidth (Tbps) min max 10 / 19 K. Pires, G. Simon DASH in Twitch
  • 17. Raw videos properties 240p 360p 480p 720p 1080p 0.6 0.4 0.2 0 Representation Sessions ratio 11 / 19 K. Pires, G. Simon DASH in Twitch
  • 18. Raw videos properties 240p 360p 480p 720p 1080p 0.6 0.4 0.2 0 Representation Sessions ratio 1 0.8 0.6 0.4 0.2 0 0.1 1 2 10 Video bit-rate (Mbps) CDF of the sessions 480p 11 / 19 K. Pires, G. Simon DASH in Twitch
  • 19. Raw videos properties 240p 360p 480p 720p 1080p 0.6 0.4 0.2 0 Representation Sessions ratio 1 0.8 0.6 0.4 0.2 0 0.1 1 2 10 Video bit-rate (Mbps) CDF of the sessions 480p 720p 1080p 11 / 19 K. Pires, G. Simon DASH in Twitch
  • 20. Selecting the ABR streams 12 / 19 K. Pires, G. Simon DASH in Twitch
  • 21. When to decide a stream should be transcoded At-startup Strategy transcoding decision time t1 t01 On-the-fly Strategy transcoding decision time t1 t01 transcoding traditional 13 / 19 K. Pires, G. Simon DASH in Twitch
  • 22. Strategies None : no ABR All : all streams are streamed with ABR 14 / 19 K. Pires, G. Simon DASH in Twitch
  • 23. Strategies None : no ABR All : all streams are streamed with ABR Top-50 (at-startup) : streams that were popular 14 / 19 K. Pires, G. Simon DASH in Twitch
  • 24. Strategies None : no ABR All : all streams are streamed with ABR Top-50 (at-startup) : streams that were popular Threshold-1000 (on-the-fly) immediate popularity 14 / 19 K. Pires, G. Simon DASH in Twitch
  • 25. Evaluation settings Sessions ! Twitch dataset Viewers ! DASH dataset ABR ! proportional to raw video 15 / 19 K. Pires, G. Simon DASH in Twitch
  • 26. Results : degraded viewers served 0 10 20 1 0.5 0 All None Days Viewers ratio Top Threshold 16 / 19 K. Pires, G. Simon DASH in Twitch
  • 27. Results : costs evaluation None All Top Threshold 150 100 50 0 -39% -40% +5% Strategies Total Price ($M) Delivery Transcode 17 / 19 K. Pires, G. Simon DASH in Twitch
  • 28. Conclusion 18 / 19 K. Pires, G. Simon DASH in Twitch
  • 29. Discussions preliminary work=global picture!details neglected more formal and accurate formulation more precise QoE estimation better transcoding needs model management of different hardware resources 19 / 19 K. Pires, G. Simon DASH in Twitch
  • 30. Discussions preliminary work=global picture!details neglected more formal and accurate formulation more precise QoE estimation better transcoding needs model management of different hardware resources more efficient strategies can be designed sophisticated strategies statistical approaches for popularity forecasting 19 / 19 K. Pires, G. Simon DASH in Twitch