SlideShare a Scribd company logo
1 of 14
Object Placement in Video Content
      Distribution Networks



        Mohammad Faraji, Kianoosh Mokhtarian
   Department of Electrical and Computer Engineering
                 University of Toronto

                    December 2011
Background

8 years of video content added to YouTube every day
  Terabytes a day; Petabytes a year
Trend is to further accelerate
  Higher-quality video streams (currently only 10% are HD)

Content distribution infrastructure
  Several datacentres around the world
  User request sent to closest datacentre (DNS/HTTP redirect)
Motivation

Store video files across datacentres (DCs)
  Generously replicate all videos on DCs?
  Not viable
  Growth of data volume >> storage cost
Good News from Measurement Studies

Popularity of video depends on geographical location

More than half of the time, only a fraction from the
 beginning of video is downloaded

=> Place (partial) video files in selected locations
Modeling

Input: history of user requests (video v for IP address i)
Distance of i to any of datacentres?
  Use an Internet Coordinate System (ICS)
  Delay(i, j) = Eucledian_distance[ ICS(i), ICS(j) ]
Make tracking of requests scalable
  Cluster user IPs into regions in the Eucledian space of ICS
Popularity matrix P[region, video]
Distance matrix D[region, datacentre]
Partial Video Files

First minute of video downloaded many more times
Store partial video files
  More effective caching
  Lower start-up delay
Partial popularity assumed independent of region
Download reports: (v, 1MB), (v, 2.3MB), (v, 0.5 MB), ...
  Compress into a few entries for each video (dynamic alg)
  PP[v] = (0...1MB, 100 times), (1MB...end, 50 times)
Problem Statement

Assign (part of) each video to one or more DC
  Minimize distance of video to user (region), given:
     The distance matrix D[region, datacentre]
     The expected download pattern P[region, video]
     Partial popularity PP[video]
     The storage limitation of each DC
Problem Hardness

Simpler alternatives

Store one video file on a few selected DCs
  NP-Complete (min set cover, max coverage)

Store multiple video files on one DC
  NP-Complete (knapsack)
Solution

Maintain a utility matirx U[v, d]
  Utility of replicating "the next chunk of" video v on DC d
  Auxiliary priority queues
1. Find the highest-utility video v*:
2. Place the next chunk of v* on the best DC d*
3. Update row v* of U, and what the next chunk is for v*
Complexity: O[ (total video replicas) x
  (log[# videos] + log[# DCs] + log[max chunks/video]) ]
Evaluation (in Progress): Data

File size and length of ~200K videos from [Cheng 2010]
Distances in Internet
  Pairwise delay between 2500 nodes from [Wong 2005]
Video popularities
  Global: Zipf-distributed (as repeatedly reported)
  Local: synthetic
Partial video popularities
  Generated according to [Qiu 2010]
Evaluation (in Progress): Results

Total delay, given our placement
  Delay w/ and wo/ partial file storage
  Comparison to simple threshold based distributed caching

Running time

Estimated communication overhead
Take-Away

Benefits of storing partial video files on selected DCs

Future work
  Sevral further details for a complete working system ...
  Low-overhead collection of (sub-samples of) downloads
  Estimate near-future download patterns
  Carefully cluster users in a limited num of regions
  Solving video placement by multiple nodes
  Incremental algorithm; can't shuffle everything every night
Appendix: Previous Works

Cooperative web caching
  Hierarchical, distributed, hybrid

CDN design (various flavors)

Video caching
  On a single cache
  To optimize for VCR-like functions

More Related Content

Viewers also liked

Modelling the Effect of Packet Loss on Speech Quality
Modelling the Effect of Packet Loss on Speech QualityModelling the Effect of Packet Loss on Speech Quality
Modelling the Effect of Packet Loss on Speech Qualityadil raja
 
Download It
Download ItDownload It
Download ItVideoguy
 
Infographic flexible connectivity with hybrid network
Infographic flexible connectivity with hybrid networkInfographic flexible connectivity with hybrid network
Infographic flexible connectivity with hybrid networkOrange Business Services
 
Mobile Multimedia Comm: Technology Dimensions
Mobile Multimedia Comm: Technology DimensionsMobile Multimedia Comm: Technology Dimensions
Mobile Multimedia Comm: Technology DimensionsDr. Aloknath De
 
VoIP Security for Dummies
VoIP Security for DummiesVoIP Security for Dummies
VoIP Security for DummiesAvaya Inc.
 

Viewers also liked (10)

Modelling the Effect of Packet Loss on Speech Quality
Modelling the Effect of Packet Loss on Speech QualityModelling the Effect of Packet Loss on Speech Quality
Modelling the Effect of Packet Loss on Speech Quality
 
Download It
Download ItDownload It
Download It
 
Infographic flexible connectivity with hybrid network
Infographic flexible connectivity with hybrid networkInfographic flexible connectivity with hybrid network
Infographic flexible connectivity with hybrid network
 
EVOLVE'14 | Maximize | Gretchen Sizer | County of San Diego
EVOLVE'14 | Maximize | Gretchen Sizer | County of San DiegoEVOLVE'14 | Maximize | Gretchen Sizer | County of San Diego
EVOLVE'14 | Maximize | Gretchen Sizer | County of San Diego
 
Mobile Multimedia Comm: Technology Dimensions
Mobile Multimedia Comm: Technology DimensionsMobile Multimedia Comm: Technology Dimensions
Mobile Multimedia Comm: Technology Dimensions
 
H.323 protocol
H.323 protocolH.323 protocol
H.323 protocol
 
E-Commerce
E-CommerceE-Commerce
E-Commerce
 
Ip telephony
Ip telephonyIp telephony
Ip telephony
 
Internet Telephony
Internet TelephonyInternet Telephony
Internet Telephony
 
VoIP Security for Dummies
VoIP Security for DummiesVoIP Security for Dummies
VoIP Security for Dummies
 

Similar to Object Placement in Video Content Distribution Networks

Paper id 28201439
Paper id 28201439Paper id 28201439
Paper id 28201439IJRAT
 
Video Streaming Ali Saman Tosun
Video Streaming Ali Saman TosunVideo Streaming Ali Saman Tosun
Video Streaming Ali Saman TosunVideoguy
 
Chapter 07
Chapter 07Chapter 07
Chapter 07 Google
 
A Benchmark to Evaluate Mobile Video Upload to Cloud Infrastructures
A Benchmark to Evaluate Mobile Video Upload to Cloud InfrastructuresA Benchmark to Evaluate Mobile Video Upload to Cloud Infrastructures
A Benchmark to Evaluate Mobile Video Upload to Cloud InfrastructuresUniversity of Southern California
 
Accessible Video in The Enterprise
Accessible Video in The Enterprise Accessible Video in The Enterprise
Accessible Video in The Enterprise John Foliot
 
PSU COE Streaming Video Offering
PSU COE Streaming Video OfferingPSU COE Streaming Video Offering
PSU COE Streaming Video OfferingVideoguy
 
Resume-LIN-en-2014
Resume-LIN-en-2014Resume-LIN-en-2014
Resume-LIN-en-2014lin xianjin
 
Resume-LIN-en-2014
Resume-LIN-en-2014Resume-LIN-en-2014
Resume-LIN-en-2014lin xianjin
 
EQR Reporting: Rails + Amazon EC2
EQR Reporting:  Rails + Amazon EC2EQR Reporting:  Rails + Amazon EC2
EQR Reporting: Rails + Amazon EC2jeperkins4
 
Video Streaming Compression for Wireless Multimedia Sensor Networks
Video Streaming Compression for Wireless Multimedia Sensor NetworksVideo Streaming Compression for Wireless Multimedia Sensor Networks
Video Streaming Compression for Wireless Multimedia Sensor NetworksIOSR Journals
 
Audio/Video Streaming over 802.11
Audio/Video Streaming over 802.11Audio/Video Streaming over 802.11
Audio/Video Streaming over 802.11Videoguy
 
Video Server
Video ServerVideo Server
Video Servernnmaurya
 
PEER-TO-PEER INTERACTIVE VOD STREAMING USING NCECD
PEER-TO-PEER INTERACTIVE VOD  STREAMING  USING NCECDPEER-TO-PEER INTERACTIVE VOD  STREAMING  USING NCECD
PEER-TO-PEER INTERACTIVE VOD STREAMING USING NCECDJayanthGubbi
 
Delay Analysis of Layered Video Caching in Crowdsourced Heterogeneous Wireles...
Delay Analysis of Layered Video Caching in Crowdsourced Heterogeneous Wireles...Delay Analysis of Layered Video Caching in Crowdsourced Heterogeneous Wireles...
Delay Analysis of Layered Video Caching in Crowdsourced Heterogeneous Wireles...Behrouz Jedari
 
Videostream compression in iOS
Videostream compression in iOSVideostream compression in iOS
Videostream compression in iOS*instinctools
 
Metrics towards enterprise readiness of unikernels
Metrics towards enterprise readiness of unikernelsMetrics towards enterprise readiness of unikernels
Metrics towards enterprise readiness of unikernelsMadhuri Yechuri
 
CAVE/RC-to-street
CAVE/RC-to-street CAVE/RC-to-street
CAVE/RC-to-street Videoguy
 
ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...
ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...
ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...Anand Bhojan
 
01_Introduction.pdf.pdf
01_Introduction.pdf.pdf01_Introduction.pdf.pdf
01_Introduction.pdf.pdfWidedMiled2
 

Similar to Object Placement in Video Content Distribution Networks (20)

060320 mmtf presentation
060320 mmtf presentation060320 mmtf presentation
060320 mmtf presentation
 
Paper id 28201439
Paper id 28201439Paper id 28201439
Paper id 28201439
 
Video Streaming Ali Saman Tosun
Video Streaming Ali Saman TosunVideo Streaming Ali Saman Tosun
Video Streaming Ali Saman Tosun
 
Chapter 07
Chapter 07Chapter 07
Chapter 07
 
A Benchmark to Evaluate Mobile Video Upload to Cloud Infrastructures
A Benchmark to Evaluate Mobile Video Upload to Cloud InfrastructuresA Benchmark to Evaluate Mobile Video Upload to Cloud Infrastructures
A Benchmark to Evaluate Mobile Video Upload to Cloud Infrastructures
 
Accessible Video in The Enterprise
Accessible Video in The Enterprise Accessible Video in The Enterprise
Accessible Video in The Enterprise
 
PSU COE Streaming Video Offering
PSU COE Streaming Video OfferingPSU COE Streaming Video Offering
PSU COE Streaming Video Offering
 
Resume-LIN-en-2014
Resume-LIN-en-2014Resume-LIN-en-2014
Resume-LIN-en-2014
 
Resume-LIN-en-2014
Resume-LIN-en-2014Resume-LIN-en-2014
Resume-LIN-en-2014
 
EQR Reporting: Rails + Amazon EC2
EQR Reporting:  Rails + Amazon EC2EQR Reporting:  Rails + Amazon EC2
EQR Reporting: Rails + Amazon EC2
 
Video Streaming Compression for Wireless Multimedia Sensor Networks
Video Streaming Compression for Wireless Multimedia Sensor NetworksVideo Streaming Compression for Wireless Multimedia Sensor Networks
Video Streaming Compression for Wireless Multimedia Sensor Networks
 
Audio/Video Streaming over 802.11
Audio/Video Streaming over 802.11Audio/Video Streaming over 802.11
Audio/Video Streaming over 802.11
 
Video Server
Video ServerVideo Server
Video Server
 
PEER-TO-PEER INTERACTIVE VOD STREAMING USING NCECD
PEER-TO-PEER INTERACTIVE VOD  STREAMING  USING NCECDPEER-TO-PEER INTERACTIVE VOD  STREAMING  USING NCECD
PEER-TO-PEER INTERACTIVE VOD STREAMING USING NCECD
 
Delay Analysis of Layered Video Caching in Crowdsourced Heterogeneous Wireles...
Delay Analysis of Layered Video Caching in Crowdsourced Heterogeneous Wireles...Delay Analysis of Layered Video Caching in Crowdsourced Heterogeneous Wireles...
Delay Analysis of Layered Video Caching in Crowdsourced Heterogeneous Wireles...
 
Videostream compression in iOS
Videostream compression in iOSVideostream compression in iOS
Videostream compression in iOS
 
Metrics towards enterprise readiness of unikernels
Metrics towards enterprise readiness of unikernelsMetrics towards enterprise readiness of unikernels
Metrics towards enterprise readiness of unikernels
 
CAVE/RC-to-street
CAVE/RC-to-street CAVE/RC-to-street
CAVE/RC-to-street
 
ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...
ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...
ShowNTell: An easy-to-use tool for answering students’ questions with voice-o...
 
01_Introduction.pdf.pdf
01_Introduction.pdf.pdf01_Introduction.pdf.pdf
01_Introduction.pdf.pdf
 

Object Placement in Video Content Distribution Networks

  • 1. Object Placement in Video Content Distribution Networks Mohammad Faraji, Kianoosh Mokhtarian Department of Electrical and Computer Engineering University of Toronto December 2011
  • 2. Background 8 years of video content added to YouTube every day Terabytes a day; Petabytes a year Trend is to further accelerate Higher-quality video streams (currently only 10% are HD) Content distribution infrastructure Several datacentres around the world User request sent to closest datacentre (DNS/HTTP redirect)
  • 3. Motivation Store video files across datacentres (DCs) Generously replicate all videos on DCs? Not viable Growth of data volume >> storage cost
  • 4. Good News from Measurement Studies Popularity of video depends on geographical location More than half of the time, only a fraction from the beginning of video is downloaded => Place (partial) video files in selected locations
  • 5. Modeling Input: history of user requests (video v for IP address i) Distance of i to any of datacentres? Use an Internet Coordinate System (ICS) Delay(i, j) = Eucledian_distance[ ICS(i), ICS(j) ] Make tracking of requests scalable Cluster user IPs into regions in the Eucledian space of ICS Popularity matrix P[region, video] Distance matrix D[region, datacentre]
  • 6. Partial Video Files First minute of video downloaded many more times Store partial video files More effective caching Lower start-up delay Partial popularity assumed independent of region Download reports: (v, 1MB), (v, 2.3MB), (v, 0.5 MB), ... Compress into a few entries for each video (dynamic alg) PP[v] = (0...1MB, 100 times), (1MB...end, 50 times)
  • 7. Problem Statement Assign (part of) each video to one or more DC Minimize distance of video to user (region), given: The distance matrix D[region, datacentre] The expected download pattern P[region, video] Partial popularity PP[video] The storage limitation of each DC
  • 8. Problem Hardness Simpler alternatives Store one video file on a few selected DCs NP-Complete (min set cover, max coverage) Store multiple video files on one DC NP-Complete (knapsack)
  • 9. Solution Maintain a utility matirx U[v, d] Utility of replicating "the next chunk of" video v on DC d Auxiliary priority queues 1. Find the highest-utility video v*: 2. Place the next chunk of v* on the best DC d* 3. Update row v* of U, and what the next chunk is for v* Complexity: O[ (total video replicas) x (log[# videos] + log[# DCs] + log[max chunks/video]) ]
  • 10. Evaluation (in Progress): Data File size and length of ~200K videos from [Cheng 2010] Distances in Internet Pairwise delay between 2500 nodes from [Wong 2005] Video popularities Global: Zipf-distributed (as repeatedly reported) Local: synthetic Partial video popularities Generated according to [Qiu 2010]
  • 11. Evaluation (in Progress): Results Total delay, given our placement Delay w/ and wo/ partial file storage Comparison to simple threshold based distributed caching Running time Estimated communication overhead
  • 12. Take-Away Benefits of storing partial video files on selected DCs Future work Sevral further details for a complete working system ... Low-overhead collection of (sub-samples of) downloads Estimate near-future download patterns Carefully cluster users in a limited num of regions Solving video placement by multiple nodes Incremental algorithm; can't shuffle everything every night
  • 13.
  • 14. Appendix: Previous Works Cooperative web caching Hierarchical, distributed, hybrid CDN design (various flavors) Video caching On a single cache To optimize for VCR-like functions

Editor's Notes

  1. We have seen a lot of works on how to build and maintain datacentres. Our work is about utilizing datacentres for a large scale CDN.
  2. There are petabytes of video files to store on the Dcs. Can't replicate everything on every DC; needs to build a whole new DC every year! Data volume is increasing at a much faster pace than the rate of storage cost decreasing.
  3. Interesting potentials that we can leverage Previous measurements on YouTube report that ...