SlideShare una empresa de Scribd logo
1 de 19
A PROXY EFFECT ANALYSIS AND FAIR
  ADAPTATION ALGORITHM FOR MULTIPLE
COMPETING DYNAMIC ADAPTIVE STREAMING
         OVER HTTP CLIENTS
                       Christopher Mueller, Stefan Lederer and Christian Timmerer
               Alpen-Adria Universität Klagenfurt (AAU)  Faculty of Technical Sciences (TEWI)
                Institute of Information Technology (ITEC)  Multimedia Communication (MMC)


                                             VCIP12 - 29-11-2012

 Christopher Mueller                   Proxy Effect Analysis and Fair Adaptation                 1
OUTLINE
 Introduction
 Proxy Effect Analysis
 Experiments with multiple competing dynamic HTTP streaming
    clients
       Testbed & Methodology
       Microsoft Smooth Streaming
       MPEG-DASH
       Fair Adaptation

 Conclusion


 Christopher Mueller      Proxy Effect Analysis and Fair Adaptation   2
DYNAMIC ADAPTIVE STREAMING
              OVER HTTP – IN A NUTSHELL
Multiple Quality Levels                                            Varying Bandwidth Conditions




      Dynamic adaption to the network conditions
                                                            Selects the appropriate
      Reuse of existing Internet infrastructure              segments for each
      Logic is located at the client side                         timepoint
      Has the potential to play a major role in future networks
       optimizations and problems analysis's are crucial
     Christopher Mueller      Proxy Effect Analysis and Fair Adaptation                  3
MOTIVATION
 Real-Time Entertainment is currently accounting for more than 50% of
  the whole internet traffic
       This traffic still grows and HTTP is one of the major protocols
       DASH has the potential to play a significant role in future networks
   Large scale scenarios are realistic use cases in the next few years
 Proxy as a caching element
       Widely deployed in the internet
       Reduces upstream and downstream and therefore costs
       Almost every ISP or bigger institution e.g., university’s, companies etc.,
        are using proxy servers
       Could fully transparent intercept the connection between the client and
        the server
   It could be assumed that nearly each HTTP connection is intercepted
   by a proxy server

 Christopher Mueller          Proxy Effect Analysis and Fair Adaptation    4
PROXY EFFECT ANALYSIS
 Potential general problems and issues for large scale dynamic
  streaming over HTTP scenarios
       Clients are not aware of the network infrastructure
       Clients are not aware of other clients
       Uncontrolled distribution of media content over caching proxies
       Usually only a portion of the content is cached
       No deep inspection of throughput measurements to specific
        segments
    A client could make an unfavorable adaptation decision
    A client could negatively influence other clients


 Christopher Mueller        Proxy Effect Analysis and Fair Adaptation   5
PROXY EFFECT ANALYSIS - EXAMPLE
 HTTP Server provides content with two quality levels
       Base Quality                   5 Mbps
       Enhancement Quality            7 Mbps
 The bottleneck bandwidth is restricted to 8 Mbps
 Client 1 could only select the base quality due to the restricted
  bandwidth
 Client 2 could potentially select the enhancement quality
  without client 1                                        6 Mbps

                                                                                                 Client 1
                                                                     8 Mbps
                                                                                             2
                                                                                      Client 1 Bandwidth
                                              HTTP Server                     Proxy

                                             Bottleneck Bandwidth                      8 Mbps

                                                                                                 Client 2
 Christopher Mueller     Proxy Effect Analysis and Fair Adaptation                         6
PROXY EFFECT ANALYSIS - EXAMPLE
 Client 1 measures an average throughput of 6 Mbps and uses
  the base quality which will be cached at the proxy
 Client 2 will also use the base quality at the beginning and
  measure and average throughput of 8 Mbps
  Client 2 will switch to the enhancement quality and the proxy
  has to host two connections to the content server which will
  be approx. shared in fair way (4 Mbps)
                                                                                    6 Mbps

                                                                                             Client 1
                                                                   8 Mbps

                                            HTTP Server                     Proxy

                                                                                    8 Mbps

                                                                                             Client 2
 Christopher Mueller   Proxy Effect Analysis and Fair Adaptation                        7
PROXY EFFECT ANALYSIS - EXAMPLE
 Both clients will now measure an effective available
  bandwidth of 4 Mbps
  Client 2 will switch back to base quality and everything starts
  from the beginning
 Problems
       Self caused frequent quality switching
       Potential unsmooth playback due to unfavorable adaptation
                                                                                        6 Mbps

                                                                                                 Client 1
                                                                       8 Mbps

                                                HTTP Server                     Proxy

                                                                                        8 Mbps

                                                                                                 Client 2
 Christopher Mueller       Proxy Effect Analysis and Fair Adaptation                        8
METHODOLOGY AND EXPERIMENTAL
            SETUP
 All experiments have been consistently performed with the
  same settings:
       Content: Big Buck Bunny encoded with x264 at two bitrates e.g.,
        700 kbps and 1300 kbps
       Proxy: Squid open source proxy in transparent mode
       Shaping: Linux traffic control and hierarchal token bucket system
 The server and client components differ only between the
  MPEG-DASH and MSS experiment
                              Same content for all experiments




                       Stays the same over all experiments
 Christopher Mueller                  Proxy Effect Analysis and Fair Adaptation   9
MICROSOFT SMOOTH STREAMING
 Server and Client Setup
       Server is based on Microsoft Windows Server 2008
       The clients are based on Microsoft Windows 7 and Silverlight
 Network Setup
       Bottleneck bandwidth has been set to 2200 kbps
       Client 1 bandwidth restricted at 1100 kbps                     Client 1 Bandwidth
                                                                          Windows 7
       Client 2 bandwidth restricted at 2200 kbps
                                               Windows Server 2008




                             700 kbps Bottleneck Bandwidth
                             1300 kbps                  Client 2 Bandwidth
                                                             Windows 7
 Christopher Mueller       Proxy Effect Analysis and Fair Adaptation           10
MICROSOFT SMOOTH STREAMING




 The MSS client behaves exactly like assumed
       Frequent self caused quality level switching
       Unsmooth playback at the beginning

 Christopher Mueller        Proxy Effect Analysis and Fair Adaptation   11
MPEG-DASH
 Another Verification has been performed with our MPEG-
  DASH implementation
       Uses HTTP/1.1 features e.g., persistent connections and pipelining
        More efficient than MSS, therefore network conditions has been
        modified. Otherwise both clients were able to stream both qualities
        smooth in parallel.

  700 kbps
                                                                        Client 2 Bandwidth
                                                                        Client 1 Bandwidth
  1300 kbps

                        Bottleneck Bandwidth

 The adaptation logic selects individual segments based on
  throughput measurements of previous segments

 Christopher Mueller        Proxy Effect Analysis and Fair Adaptation           12
MPEG-DASH




 Same problems like the MSS client
       More frequent quality level switching due to the more aggressive
        adaptation
       Lower buffer level and unsmooth playback
 Christopher Mueller       Proxy Effect Analysis and Fair Adaptation   13
FAIR ADAPTATION
 Tries to address this problems and consider the locality of
  segments
 MPEG-DASH VLC plug-in has been
  extended for Fair Adaptation
 Uses a probe function to verify the
  measured throughput
       Non-Cacheable object on the server
       The proxy server could actively modify
        the MPD
       The proxy could provide a service
       Client could download the first few bytes
        or a random byte range



 Christopher Mueller        Proxy Effect Analysis and Fair Adaptation   14
FAIR ADAPTATION




 Probe in combination with an exponential backoff depicted
  with green vertical lines
 No frequent self caused quality switching
 More stable buffer level
 Christopher Mueller      Proxy Effect Analysis and Fair Adaptation   15
FAIR ADAPTATION
 Client problems
       Self caused quality switching has been eliminated
       No rebuffering events or stalls have been occurred
       Increases average buffer fill state
       Increases stability of the session
       Other clients will not be influenced
 Bottleneck performance
       Maximizes the representation reuse
       Minimizes the used bottleneck bandwidth
 Network and Standard
       No changes on the network side are needed
       No changes in the standard are needed
 Christopher Mueller        Proxy Effect Analysis and Fair Adaptation   16
CONCLUSION
 Uncontrolled distribution of content over caching proxies could
  negatively influence the streaming performance of several
  clients
 Experimentally evaluated the effects of caching proxies on
  dynamic adaptive streaming over HTTP
       Assumptions have been proven with Microsoft Smooth Streaming
        and MPEG-DASH
       The negative effects depend on the streaming logic and the client
        constellation
 A Fair Adaptation Algorithm has been proposed that reduces
  the negative effects
       Needs no changes on the standard or network side
 Further work will include the dynamic case and more clients
 Christopher Mueller       Proxy Effect Analysis and Fair Adaptation   17
THANK YOU FOR YOUR ATTENTION



                  … questions, comments, etc. are welcome …


                           Christopher Mueller | dash.itec.aau.at
         Alpen-Adria Universität Klagenfurt (AAU)  Faculty of Technical Sciences (TEWI)
         Institute of Information Technology (ITEC)  Multimedia Communication (MMC)




Christopher Mueller               Proxy Effect Analysis and Fair Adaptation           18
ALGORITHM IN DETAIL




Christopher Mueller        Proxy Effect Analysis and Fair Adaptation   19

Más contenido relacionado

La actualidad más candente

Adaptive Media Streaming over Emerging Protocols
Adaptive Media Streaming over Emerging ProtocolsAdaptive Media Streaming over Emerging Protocols
Adaptive Media Streaming over Emerging ProtocolsAlpen-Adria-Universität
 
Standards' Perspective - MPEG DASH overview and related efforts
Standards' Perspective - MPEG DASH overview and related effortsStandards' Perspective - MPEG DASH overview and related efforts
Standards' Perspective - MPEG DASH overview and related effortsIMTC
 
Emulation of Dynamic Adaptive Streaming over HTTP with Mininet
Emulation of Dynamic Adaptive Streaming over HTTP with MininetEmulation of Dynamic Adaptive Streaming over HTTP with Mininet
Emulation of Dynamic Adaptive Streaming over HTTP with MininetAnatoliy Zabrovskiy
 
Dynamic Adaptive Streaming over HTTP (DASH)
Dynamic Adaptive Streaming over HTTP (DASH)Dynamic Adaptive Streaming over HTTP (DASH)
Dynamic Adaptive Streaming over HTTP (DASH)Alpen-Adria-Universität
 
Ebu mpeg dash-webinar043
Ebu mpeg dash-webinar043Ebu mpeg dash-webinar043
Ebu mpeg dash-webinar043mc_killah
 
Using DASH and MPEG-2 TS
Using DASH and MPEG-2 TSUsing DASH and MPEG-2 TS
Using DASH and MPEG-2 TSAlex Giladi
 
Edge 2014: MPEG DASH – Tomorrow's Format Today
Edge 2014: MPEG DASH – Tomorrow's Format TodayEdge 2014: MPEG DASH – Tomorrow's Format Today
Edge 2014: MPEG DASH – Tomorrow's Format TodayAkamai Technologies
 
MPEG DASH White Paper
MPEG DASH White PaperMPEG DASH White Paper
MPEG DASH White Paperidrajeev
 
A Seamless Web Integration of Adaptive HTTP Streaming
A Seamless Web Integration of Adaptive HTTP StreamingA Seamless Web Integration of Adaptive HTTP Streaming
A Seamless Web Integration of Adaptive HTTP StreamingAlpen-Adria-Universität
 
An Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular Environments
An Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular EnvironmentsAn Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular Environments
An Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular EnvironmentsAlpen-Adria-Universität
 
Dynamic Adaptive Streaming over HTTP: From Content Creation to Consumption
Dynamic Adaptive Streaming over HTTP: From Content Creation to ConsumptionDynamic Adaptive Streaming over HTTP: From Content Creation to Consumption
Dynamic Adaptive Streaming over HTTP: From Content Creation to ConsumptionAlpen-Adria-Universität
 
MPEG-DASH Reference Software and Conformance
MPEG-DASH Reference Software and ConformanceMPEG-DASH Reference Software and Conformance
MPEG-DASH Reference Software and ConformanceAlpen-Adria-Universität
 
Implementation and Performance Analysis of a UDP Binding for SOAP
Implementation and Performance Analysis of a UDP Binding for SOAPImplementation and Performance Analysis of a UDP Binding for SOAP
Implementation and Performance Analysis of a UDP Binding for SOAPDr. Fahad Aijaz
 
Mini proj ii sdn video communication
Mini proj ii   sdn video communicationMini proj ii   sdn video communication
Mini proj ii sdn video communicationHaowei Jiang
 
Different Types of Live Video Streaming Protocols and the Way They Work
Different Types of Live Video Streaming Protocols and the Way They WorkDifferent Types of Live Video Streaming Protocols and the Way They Work
Different Types of Live Video Streaming Protocols and the Way They WorkChetu
 
HTTP Adaptive Streaming State of the Art and Challenges Ahead
HTTP Adaptive StreamingState of the Art and Challenges AheadHTTP Adaptive StreamingState of the Art and Challenges Ahead
HTTP Adaptive Streaming State of the Art and Challenges AheadAlpen-Adria-Universität
 

La actualidad más candente (20)

HTTP Streaming of MPEG Media
HTTP Streaming of MPEG MediaHTTP Streaming of MPEG Media
HTTP Streaming of MPEG Media
 
Adaptive Media Streaming over Emerging Protocols
Adaptive Media Streaming over Emerging ProtocolsAdaptive Media Streaming over Emerging Protocols
Adaptive Media Streaming over Emerging Protocols
 
Standards' Perspective - MPEG DASH overview and related efforts
Standards' Perspective - MPEG DASH overview and related effortsStandards' Perspective - MPEG DASH overview and related efforts
Standards' Perspective - MPEG DASH overview and related efforts
 
Emulation of Dynamic Adaptive Streaming over HTTP with Mininet
Emulation of Dynamic Adaptive Streaming over HTTP with MininetEmulation of Dynamic Adaptive Streaming over HTTP with Mininet
Emulation of Dynamic Adaptive Streaming over HTTP with Mininet
 
MPEG-DASH open source tools and cloud services
MPEG-DASH open source tools and cloud servicesMPEG-DASH open source tools and cloud services
MPEG-DASH open source tools and cloud services
 
Dynamic Adaptive Streaming over HTTP (DASH)
Dynamic Adaptive Streaming over HTTP (DASH)Dynamic Adaptive Streaming over HTTP (DASH)
Dynamic Adaptive Streaming over HTTP (DASH)
 
Ebu mpeg dash-webinar043
Ebu mpeg dash-webinar043Ebu mpeg dash-webinar043
Ebu mpeg dash-webinar043
 
Using DASH and MPEG-2 TS
Using DASH and MPEG-2 TSUsing DASH and MPEG-2 TS
Using DASH and MPEG-2 TS
 
Edge 2014: MPEG DASH – Tomorrow's Format Today
Edge 2014: MPEG DASH – Tomorrow's Format TodayEdge 2014: MPEG DASH – Tomorrow's Format Today
Edge 2014: MPEG DASH – Tomorrow's Format Today
 
Distributed DASH Dataset
Distributed DASH DatasetDistributed DASH Dataset
Distributed DASH Dataset
 
Dynamic Adaptive Streaming over HTTP Dataset
Dynamic Adaptive Streaming over HTTP DatasetDynamic Adaptive Streaming over HTTP Dataset
Dynamic Adaptive Streaming over HTTP Dataset
 
MPEG DASH White Paper
MPEG DASH White PaperMPEG DASH White Paper
MPEG DASH White Paper
 
A Seamless Web Integration of Adaptive HTTP Streaming
A Seamless Web Integration of Adaptive HTTP StreamingA Seamless Web Integration of Adaptive HTTP Streaming
A Seamless Web Integration of Adaptive HTTP Streaming
 
An Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular Environments
An Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular EnvironmentsAn Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular Environments
An Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular Environments
 
Dynamic Adaptive Streaming over HTTP: From Content Creation to Consumption
Dynamic Adaptive Streaming over HTTP: From Content Creation to ConsumptionDynamic Adaptive Streaming over HTTP: From Content Creation to Consumption
Dynamic Adaptive Streaming over HTTP: From Content Creation to Consumption
 
MPEG-DASH Reference Software and Conformance
MPEG-DASH Reference Software and ConformanceMPEG-DASH Reference Software and Conformance
MPEG-DASH Reference Software and Conformance
 
Implementation and Performance Analysis of a UDP Binding for SOAP
Implementation and Performance Analysis of a UDP Binding for SOAPImplementation and Performance Analysis of a UDP Binding for SOAP
Implementation and Performance Analysis of a UDP Binding for SOAP
 
Mini proj ii sdn video communication
Mini proj ii   sdn video communicationMini proj ii   sdn video communication
Mini proj ii sdn video communication
 
Different Types of Live Video Streaming Protocols and the Way They Work
Different Types of Live Video Streaming Protocols and the Way They WorkDifferent Types of Live Video Streaming Protocols and the Way They Work
Different Types of Live Video Streaming Protocols and the Way They Work
 
HTTP Adaptive Streaming State of the Art and Challenges Ahead
HTTP Adaptive StreamingState of the Art and Challenges AheadHTTP Adaptive StreamingState of the Art and Challenges Ahead
HTTP Adaptive Streaming State of the Art and Challenges Ahead
 

Similar a A PROXY EFFECT ANALYIS AND FAIR ADATPATION ALGORITHM FOR MULTIPLE COMPETING DYNAMIC ADAPTIVE STREAMING OVER HTTP CLIENTS

Application Layer Throughput Control For Video Streaming over HTTP2
Application Layer Throughput Control For Video Streaming over HTTP2Application Layer Throughput Control For Video Streaming over HTTP2
Application Layer Throughput Control For Video Streaming over HTTP2Duc Nguyen
 
band width ppt
band width pptband width ppt
band width pptSai Nukala
 
口試投影片(詹智傑) Final
口試投影片(詹智傑) Final口試投影片(詹智傑) Final
口試投影片(詹智傑) Final詹智傑
 
NEW APPROACH TO DEVELOP THE MESSENGER APPLICATION: FROM CLIENTSERVER DESIGN T...
NEW APPROACH TO DEVELOP THE MESSENGER APPLICATION: FROM CLIENTSERVER DESIGN T...NEW APPROACH TO DEVELOP THE MESSENGER APPLICATION: FROM CLIENTSERVER DESIGN T...
NEW APPROACH TO DEVELOP THE MESSENGER APPLICATION: FROM CLIENTSERVER DESIGN T...cscpconf
 
Internet Path Selection on Video QoE Analysis and Improvements
Internet Path Selection on Video QoE Analysis and ImprovementsInternet Path Selection on Video QoE Analysis and Improvements
Internet Path Selection on Video QoE Analysis and ImprovementsIJTET Journal
 
Providing Controlled Quality Assurance in Video Streaming ...
Providing Controlled Quality Assurance in Video Streaming ...Providing Controlled Quality Assurance in Video Streaming ...
Providing Controlled Quality Assurance in Video Streaming ...Videoguy
 
JPN1402 A Study on False Channel Condition Reporting Attacks in Wireless Ne...
JPN1402   A Study on False Channel Condition Reporting Attacks in Wireless Ne...JPN1402   A Study on False Channel Condition Reporting Attacks in Wireless Ne...
JPN1402 A Study on False Channel Condition Reporting Attacks in Wireless Ne...chennaijp
 
Chaining Algorithm and Protocol for Peer-to-Peer Streaming Video on Demand Sy...
Chaining Algorithm and Protocol for Peer-to-Peer Streaming Video on Demand Sy...Chaining Algorithm and Protocol for Peer-to-Peer Streaming Video on Demand Sy...
Chaining Algorithm and Protocol for Peer-to-Peer Streaming Video on Demand Sy...ijwmn
 
Bandwidth estimation for ieee 802
Bandwidth estimation for ieee 802Bandwidth estimation for ieee 802
Bandwidth estimation for ieee 802Mumbai Academisc
 
Managing the Mobile Device Wave
Managing the Mobile Device WaveManaging the Mobile Device Wave
Managing the Mobile Device WaveCisco Canada
 
A Design Pattern for Reliable HTTP-Based Applications
A Design Pattern for Reliable HTTP-Based ApplicationsA Design Pattern for Reliable HTTP-Based Applications
A Design Pattern for Reliable HTTP-Based Applicationsrajkumar raj
 
Quality adaptive p2 p streaming using svc ricky 9876059 - multimedia data c...
Quality adaptive p2 p streaming using svc   ricky 9876059 - multimedia data c...Quality adaptive p2 p streaming using svc   ricky 9876059 - multimedia data c...
Quality adaptive p2 p streaming using svc ricky 9876059 - multimedia data c...Ricky Christanto
 
IT Solutions Provider in Kosovo uses Bandwidth monitoring, NetFlow Analyzer
IT Solutions Provider in Kosovo uses Bandwidth monitoring, NetFlow AnalyzerIT Solutions Provider in Kosovo uses Bandwidth monitoring, NetFlow Analyzer
IT Solutions Provider in Kosovo uses Bandwidth monitoring, NetFlow AnalyzerManageEngine, Zoho Corporation
 
dynamic media streaming over wireless and ip networks
dynamic media streaming over wireless and ip networksdynamic media streaming over wireless and ip networks
dynamic media streaming over wireless and ip networksNaveen Dubey
 
Performance Evaluation of XMPP on the Web
Performance Evaluation of XMPP on the WebPerformance Evaluation of XMPP on the Web
Performance Evaluation of XMPP on the WebMarkku Laine
 
juniper qos.ppt
juniper qos.pptjuniper qos.ppt
juniper qos.pptSunnyLai23
 
Managing the Mobile Device Wave for Enterpise Wireless Networks: Best Practices
Managing the Mobile Device Wave for Enterpise Wireless Networks: Best PracticesManaging the Mobile Device Wave for Enterpise Wireless Networks: Best Practices
Managing the Mobile Device Wave for Enterpise Wireless Networks: Best PracticesCisco Mobility
 
The Effect of Seeking Operation on QoE of HTTP Adaptive Streaming Services
The Effect of Seeking Operation on QoE of HTTP Adaptive Streaming ServicesThe Effect of Seeking Operation on QoE of HTTP Adaptive Streaming Services
The Effect of Seeking Operation on QoE of HTTP Adaptive Streaming ServicesIJCNCJournal
 

Similar a A PROXY EFFECT ANALYIS AND FAIR ADATPATION ALGORITHM FOR MULTIPLE COMPETING DYNAMIC ADAPTIVE STREAMING OVER HTTP CLIENTS (20)

Application Layer Throughput Control For Video Streaming over HTTP2
Application Layer Throughput Control For Video Streaming over HTTP2Application Layer Throughput Control For Video Streaming over HTTP2
Application Layer Throughput Control For Video Streaming over HTTP2
 
band width ppt
band width pptband width ppt
band width ppt
 
口試投影片(詹智傑) Final
口試投影片(詹智傑) Final口試投影片(詹智傑) Final
口試投影片(詹智傑) Final
 
NEW APPROACH TO DEVELOP THE MESSENGER APPLICATION: FROM CLIENTSERVER DESIGN T...
NEW APPROACH TO DEVELOP THE MESSENGER APPLICATION: FROM CLIENTSERVER DESIGN T...NEW APPROACH TO DEVELOP THE MESSENGER APPLICATION: FROM CLIENTSERVER DESIGN T...
NEW APPROACH TO DEVELOP THE MESSENGER APPLICATION: FROM CLIENTSERVER DESIGN T...
 
Internet Path Selection on Video QoE Analysis and Improvements
Internet Path Selection on Video QoE Analysis and ImprovementsInternet Path Selection on Video QoE Analysis and Improvements
Internet Path Selection on Video QoE Analysis and Improvements
 
Providing Controlled Quality Assurance in Video Streaming ...
Providing Controlled Quality Assurance in Video Streaming ...Providing Controlled Quality Assurance in Video Streaming ...
Providing Controlled Quality Assurance in Video Streaming ...
 
JPN1402 A Study on False Channel Condition Reporting Attacks in Wireless Ne...
JPN1402   A Study on False Channel Condition Reporting Attacks in Wireless Ne...JPN1402   A Study on False Channel Condition Reporting Attacks in Wireless Ne...
JPN1402 A Study on False Channel Condition Reporting Attacks in Wireless Ne...
 
Chaining Algorithm and Protocol for Peer-to-Peer Streaming Video on Demand Sy...
Chaining Algorithm and Protocol for Peer-to-Peer Streaming Video on Demand Sy...Chaining Algorithm and Protocol for Peer-to-Peer Streaming Video on Demand Sy...
Chaining Algorithm and Protocol for Peer-to-Peer Streaming Video on Demand Sy...
 
Bandwidth estimation for ieee 802
Bandwidth estimation for ieee 802Bandwidth estimation for ieee 802
Bandwidth estimation for ieee 802
 
Managing the Mobile Device Wave
Managing the Mobile Device WaveManaging the Mobile Device Wave
Managing the Mobile Device Wave
 
A Design Pattern for Reliable HTTP-Based Applications
A Design Pattern for Reliable HTTP-Based ApplicationsA Design Pattern for Reliable HTTP-Based Applications
A Design Pattern for Reliable HTTP-Based Applications
 
Quality adaptive p2 p streaming using svc ricky 9876059 - multimedia data c...
Quality adaptive p2 p streaming using svc   ricky 9876059 - multimedia data c...Quality adaptive p2 p streaming using svc   ricky 9876059 - multimedia data c...
Quality adaptive p2 p streaming using svc ricky 9876059 - multimedia data c...
 
IT Solutions Provider in Kosovo uses Bandwidth monitoring, NetFlow Analyzer
IT Solutions Provider in Kosovo uses Bandwidth monitoring, NetFlow AnalyzerIT Solutions Provider in Kosovo uses Bandwidth monitoring, NetFlow Analyzer
IT Solutions Provider in Kosovo uses Bandwidth monitoring, NetFlow Analyzer
 
dynamic media streaming over wireless and ip networks
dynamic media streaming over wireless and ip networksdynamic media streaming over wireless and ip networks
dynamic media streaming over wireless and ip networks
 
Performance Evaluation of XMPP on the Web
Performance Evaluation of XMPP on the WebPerformance Evaluation of XMPP on the Web
Performance Evaluation of XMPP on the Web
 
juniper qos.ppt
juniper qos.pptjuniper qos.ppt
juniper qos.ppt
 
Aa
AaAa
Aa
 
Managing the Mobile Device Wave for Enterpise Wireless Networks: Best Practices
Managing the Mobile Device Wave for Enterpise Wireless Networks: Best PracticesManaging the Mobile Device Wave for Enterpise Wireless Networks: Best Practices
Managing the Mobile Device Wave for Enterpise Wireless Networks: Best Practices
 
The Effect of Seeking Operation on QoE of HTTP Adaptive Streaming Services
The Effect of Seeking Operation on QoE of HTTP Adaptive Streaming ServicesThe Effect of Seeking Operation on QoE of HTTP Adaptive Streaming Services
The Effect of Seeking Operation on QoE of HTTP Adaptive Streaming Services
 
Chapter04
Chapter04Chapter04
Chapter04
 

Último

Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingEdi Saputra
 
A Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source MilvusA Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source MilvusZilliz
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...Zilliz
 
GenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdfGenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdflior mazor
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsNanddeep Nachan
 
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...Zilliz
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoffsammart93
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Miguel Araújo
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProduct Anonymous
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processorsdebabhi2
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native ApplicationsWSO2
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobeapidays
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century educationjfdjdjcjdnsjd
 
AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024The Digital Insurer
 

Último (20)

Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
 
A Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source MilvusA Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source Milvus
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
GenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdfGenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdf
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectors
 
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
Emergent Methods: Multi-lingual narrative tracking in the news - real-time ex...
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024
 

A PROXY EFFECT ANALYIS AND FAIR ADATPATION ALGORITHM FOR MULTIPLE COMPETING DYNAMIC ADAPTIVE STREAMING OVER HTTP CLIENTS

  • 1. A PROXY EFFECT ANALYSIS AND FAIR ADAPTATION ALGORITHM FOR MULTIPLE COMPETING DYNAMIC ADAPTIVE STREAMING OVER HTTP CLIENTS Christopher Mueller, Stefan Lederer and Christian Timmerer Alpen-Adria Universität Klagenfurt (AAU)  Faculty of Technical Sciences (TEWI) Institute of Information Technology (ITEC)  Multimedia Communication (MMC) VCIP12 - 29-11-2012 Christopher Mueller Proxy Effect Analysis and Fair Adaptation 1
  • 2. OUTLINE  Introduction  Proxy Effect Analysis  Experiments with multiple competing dynamic HTTP streaming clients  Testbed & Methodology  Microsoft Smooth Streaming  MPEG-DASH  Fair Adaptation  Conclusion Christopher Mueller Proxy Effect Analysis and Fair Adaptation 2
  • 3. DYNAMIC ADAPTIVE STREAMING OVER HTTP – IN A NUTSHELL Multiple Quality Levels Varying Bandwidth Conditions  Dynamic adaption to the network conditions Selects the appropriate  Reuse of existing Internet infrastructure segments for each  Logic is located at the client side timepoint  Has the potential to play a major role in future networks optimizations and problems analysis's are crucial Christopher Mueller Proxy Effect Analysis and Fair Adaptation 3
  • 4. MOTIVATION  Real-Time Entertainment is currently accounting for more than 50% of the whole internet traffic  This traffic still grows and HTTP is one of the major protocols  DASH has the potential to play a significant role in future networks Large scale scenarios are realistic use cases in the next few years  Proxy as a caching element  Widely deployed in the internet  Reduces upstream and downstream and therefore costs  Almost every ISP or bigger institution e.g., university’s, companies etc., are using proxy servers  Could fully transparent intercept the connection between the client and the server It could be assumed that nearly each HTTP connection is intercepted by a proxy server Christopher Mueller Proxy Effect Analysis and Fair Adaptation 4
  • 5. PROXY EFFECT ANALYSIS  Potential general problems and issues for large scale dynamic streaming over HTTP scenarios  Clients are not aware of the network infrastructure  Clients are not aware of other clients  Uncontrolled distribution of media content over caching proxies  Usually only a portion of the content is cached  No deep inspection of throughput measurements to specific segments A client could make an unfavorable adaptation decision A client could negatively influence other clients Christopher Mueller Proxy Effect Analysis and Fair Adaptation 5
  • 6. PROXY EFFECT ANALYSIS - EXAMPLE  HTTP Server provides content with two quality levels  Base Quality 5 Mbps  Enhancement Quality 7 Mbps  The bottleneck bandwidth is restricted to 8 Mbps  Client 1 could only select the base quality due to the restricted bandwidth  Client 2 could potentially select the enhancement quality without client 1 6 Mbps Client 1 8 Mbps 2 Client 1 Bandwidth HTTP Server Proxy Bottleneck Bandwidth 8 Mbps Client 2 Christopher Mueller Proxy Effect Analysis and Fair Adaptation 6
  • 7. PROXY EFFECT ANALYSIS - EXAMPLE  Client 1 measures an average throughput of 6 Mbps and uses the base quality which will be cached at the proxy  Client 2 will also use the base quality at the beginning and measure and average throughput of 8 Mbps Client 2 will switch to the enhancement quality and the proxy has to host two connections to the content server which will be approx. shared in fair way (4 Mbps) 6 Mbps Client 1 8 Mbps HTTP Server Proxy 8 Mbps Client 2 Christopher Mueller Proxy Effect Analysis and Fair Adaptation 7
  • 8. PROXY EFFECT ANALYSIS - EXAMPLE  Both clients will now measure an effective available bandwidth of 4 Mbps Client 2 will switch back to base quality and everything starts from the beginning  Problems  Self caused frequent quality switching  Potential unsmooth playback due to unfavorable adaptation 6 Mbps Client 1 8 Mbps HTTP Server Proxy 8 Mbps Client 2 Christopher Mueller Proxy Effect Analysis and Fair Adaptation 8
  • 9. METHODOLOGY AND EXPERIMENTAL SETUP  All experiments have been consistently performed with the same settings:  Content: Big Buck Bunny encoded with x264 at two bitrates e.g., 700 kbps and 1300 kbps  Proxy: Squid open source proxy in transparent mode  Shaping: Linux traffic control and hierarchal token bucket system  The server and client components differ only between the MPEG-DASH and MSS experiment Same content for all experiments Stays the same over all experiments Christopher Mueller Proxy Effect Analysis and Fair Adaptation 9
  • 10. MICROSOFT SMOOTH STREAMING  Server and Client Setup  Server is based on Microsoft Windows Server 2008  The clients are based on Microsoft Windows 7 and Silverlight  Network Setup  Bottleneck bandwidth has been set to 2200 kbps  Client 1 bandwidth restricted at 1100 kbps Client 1 Bandwidth Windows 7  Client 2 bandwidth restricted at 2200 kbps Windows Server 2008 700 kbps Bottleneck Bandwidth 1300 kbps Client 2 Bandwidth Windows 7 Christopher Mueller Proxy Effect Analysis and Fair Adaptation 10
  • 11. MICROSOFT SMOOTH STREAMING  The MSS client behaves exactly like assumed  Frequent self caused quality level switching  Unsmooth playback at the beginning Christopher Mueller Proxy Effect Analysis and Fair Adaptation 11
  • 12. MPEG-DASH  Another Verification has been performed with our MPEG- DASH implementation  Uses HTTP/1.1 features e.g., persistent connections and pipelining More efficient than MSS, therefore network conditions has been modified. Otherwise both clients were able to stream both qualities smooth in parallel. 700 kbps Client 2 Bandwidth Client 1 Bandwidth 1300 kbps Bottleneck Bandwidth  The adaptation logic selects individual segments based on throughput measurements of previous segments Christopher Mueller Proxy Effect Analysis and Fair Adaptation 12
  • 13. MPEG-DASH  Same problems like the MSS client  More frequent quality level switching due to the more aggressive adaptation  Lower buffer level and unsmooth playback Christopher Mueller Proxy Effect Analysis and Fair Adaptation 13
  • 14. FAIR ADAPTATION  Tries to address this problems and consider the locality of segments  MPEG-DASH VLC plug-in has been extended for Fair Adaptation  Uses a probe function to verify the measured throughput  Non-Cacheable object on the server  The proxy server could actively modify the MPD  The proxy could provide a service  Client could download the first few bytes or a random byte range Christopher Mueller Proxy Effect Analysis and Fair Adaptation 14
  • 15. FAIR ADAPTATION  Probe in combination with an exponential backoff depicted with green vertical lines  No frequent self caused quality switching  More stable buffer level Christopher Mueller Proxy Effect Analysis and Fair Adaptation 15
  • 16. FAIR ADAPTATION  Client problems  Self caused quality switching has been eliminated  No rebuffering events or stalls have been occurred  Increases average buffer fill state  Increases stability of the session  Other clients will not be influenced  Bottleneck performance  Maximizes the representation reuse  Minimizes the used bottleneck bandwidth  Network and Standard  No changes on the network side are needed  No changes in the standard are needed Christopher Mueller Proxy Effect Analysis and Fair Adaptation 16
  • 17. CONCLUSION  Uncontrolled distribution of content over caching proxies could negatively influence the streaming performance of several clients  Experimentally evaluated the effects of caching proxies on dynamic adaptive streaming over HTTP  Assumptions have been proven with Microsoft Smooth Streaming and MPEG-DASH  The negative effects depend on the streaming logic and the client constellation  A Fair Adaptation Algorithm has been proposed that reduces the negative effects  Needs no changes on the standard or network side  Further work will include the dynamic case and more clients Christopher Mueller Proxy Effect Analysis and Fair Adaptation 17
  • 18. THANK YOU FOR YOUR ATTENTION … questions, comments, etc. are welcome … Christopher Mueller | dash.itec.aau.at Alpen-Adria Universität Klagenfurt (AAU)  Faculty of Technical Sciences (TEWI) Institute of Information Technology (ITEC)  Multimedia Communication (MMC) Christopher Mueller Proxy Effect Analysis and Fair Adaptation 18
  • 19. ALGORITHM IN DETAIL Christopher Mueller Proxy Effect Analysis and Fair Adaptation 19