Measuring Decentralized Video Streaming: Background Methodology A - - PowerPoint PPT Presentation

▶
measuring decentralized video streaming
SMART_READER_LITE
LIVE PREVIEW

Measuring Decentralized Video Streaming: Background Methodology A - - PowerPoint PPT Presentation

Jun 06, 2023 650 likes •1.27k views

Motivation Measuring Decentralized Video Streaming: Background Methodology A Case Study of DTube Analysis Conclusions Trinh Viet Doan, Tat Dat Pham, Markus Oberprieler, Vaibhav Bajpai Technical University of Munich (TUM) IFIP Networking,

slide-1
SLIDE 1

Motivation Background Methodology Analysis Conclusions

Measuring Decentralized Video Streaming: A Case Study of DTube

Trinh Viet Doan, Tat Dat Pham, Markus Oberprieler, Vaibhav Bajpai

Technical University of Munich (TUM)

IFIP Networking, June 22–25, 2020

1 / 20

slide-2
SLIDE 2

Motivation Background Methodology Analysis Conclusions

Motivation

2 / 20

slide-3
SLIDE 3

Motivation Background Methodology Analysis Conclusions

Motivation

◮ Increasing concerns about consolidation in the Internet [1, 2, 3]

3 / 20

slide-4
SLIDE 4

Motivation Background Methodology Analysis Conclusions

Motivation

◮ Increasing concerns about consolidation in the Internet [1, 2, 3] ◮ Most Internet traffic: video streaming, mostly from centralized services

◮ YouTube 8.7%, Netflix 12.6% of all global downstream traffic as of 2019 [4] 3 / 20

slide-5
SLIDE 5

Motivation Background Methodology Analysis Conclusions

Motivation

◮ Increasing concerns about consolidation in the Internet [1, 2, 3] ◮ Most Internet traffic: video streaming, mostly from centralized services

◮ YouTube 8.7%, Netflix 12.6% of all global downstream traffic as of 2019 [4]

◮ Proposals of decentralized solutions to counteract centralization

3 / 20

slide-6
SLIDE 6

Motivation Background Methodology Analysis Conclusions

Motivation

◮ Increasing concerns about consolidation in the Internet [1, 2, 3] ◮ Most Internet traffic: video streaming, mostly from centralized services

◮ YouTube 8.7%, Netflix 12.6% of all global downstream traffic as of 2019 [4]

◮ Proposals of decentralized solutions to counteract centralization ◮ Decentralized video streaming:

◮ In the past: P2P video streaming around 2010 ◮ More recently: DTube (2017), PeerTube (2018), LBRY.tv (2020), ... 3 / 20

slide-7
SLIDE 7

Motivation Background Methodology Analysis Conclusions

Motivation

◮ Increasing concerns about consolidation in the Internet [1, 2, 3] ◮ Most Internet traffic: video streaming, mostly from centralized services

◮ YouTube 8.7%, Netflix 12.6% of all global downstream traffic as of 2019 [4]

◮ Proposals of decentralized solutions to counteract centralization ◮ Decentralized video streaming:

◮ In the past: P2P video streaming around 2010 ◮ More recently: DTube (2017), PeerTube (2018), LBRY.tv (2020), ...

⇒ Comparing video streaming from centralized and decentralized services using YouTube and DTube

3 / 20

slide-8
SLIDE 8

Motivation Background Methodology Analysis Conclusions

Background

4 / 20

slide-9
SLIDE 9

Motivation Background Methodology Analysis Conclusions

Background: DTube

◮ Decentralized video streaming service

5 / 20

slide-10
SLIDE 10

Motivation Background Methodology Analysis Conclusions

Background: DTube

◮ Decentralized video streaming service ◮ Leverages variety of decentralized technologies

5 / 20

slide-11
SLIDE 11

Motivation Background Methodology Analysis Conclusions

Background: DTube

◮ Decentralized video streaming service ◮ Leverages variety of decentralized technologies

◮ Interplanetary File System (IPFS) for video storage ◮ Decentralized P2P network for storage and delivery of files ◮ IPFS gateway to access IPFS content over HTTP 5 / 20

slide-12
SLIDE 12

Motivation Background Methodology Analysis Conclusions

Background: DTube

◮ Decentralized video streaming service ◮ Leverages variety of decentralized technologies

◮ Interplanetary File System (IPFS) for video storage ◮ Decentralized P2P network for storage and delivery of files ◮ IPFS gateway to access IPFS content over HTTP ◮ Steem blockchain ecosystem for user- and metadata management ◮ Token rewards to incentivize content contribution 5 / 20

slide-13
SLIDE 13

Motivation Background Methodology Analysis Conclusions

Background: DTube

◮ Decentralized video streaming service ◮ Leverages variety of decentralized technologies

◮ Interplanetary File System (IPFS) for video storage ◮ Decentralized P2P network for storage and delivery of files ◮ IPFS gateway to access IPFS content over HTTP ◮ Steem blockchain ecosystem for user- and metadata management ◮ Token rewards to incentivize content contribution

◮ Parallels to YouTube in terms of user-interaction features, user interface,

monetary incentives

5 / 20

slide-14
SLIDE 14

Motivation Background Methodology Analysis Conclusions

Background: DTube

◮ Decentralized video streaming service ◮ Leverages variety of decentralized technologies

◮ Interplanetary File System (IPFS) for video storage ◮ Decentralized P2P network for storage and delivery of files ◮ IPFS gateway to access IPFS content over HTTP ◮ Steem blockchain ecosystem for user- and metadata management ◮ Token rewards to incentivize content contribution

◮ Parallels to YouTube in terms of user-interaction features, user interface,

monetary incentives

Note: Several additions/changes to DTube’s design since beginning of study in early 2019

5 / 20

slide-15
SLIDE 15

Motivation Background Methodology Analysis Conclusions

Methodology

6 / 20

slide-16
SLIDE 16

Motivation Background Methodology Analysis Conclusions

Methodology

◮ Development of (open-source) Android application

to measure video streaming

7 / 20

slide-17
SLIDE 17

Motivation Background Methodology Analysis Conclusions

Methodology

◮ Development of (open-source) Android application

to measure video streaming

◮ Measure both YouTube and DTube with the same

framework/logic

◮ From user perspective ◮ Possibility to add other services in the future 7 / 20

slide-18
SLIDE 18

Motivation Background Methodology Analysis Conclusions

Methodology

◮ Development of (open-source) Android application

to measure video streaming

◮ Measure both YouTube and DTube with the same

framework/logic

◮ From user perspective ◮ Possibility to add other services in the future

◮ Playout of videos using ExoPlayer

7 / 20

slide-19
SLIDE 19

Motivation Background Methodology Analysis Conclusions

Methodology

◮ Development of (open-source) Android application

to measure video streaming

◮ Measure both YouTube and DTube with the same

framework/logic

◮ From user perspective ◮ Possibility to add other services in the future

◮ Playout of videos using ExoPlayer

7 / 20

slide-20
SLIDE 20

Motivation Background Methodology Analysis Conclusions

Methodology

Measurement Process

8 / 20

slide-21
SLIDE 21

Motivation Background Methodology Analysis Conclusions

Methodology

Measurement Process

  • 1. Acquire Web page URLs for n videos from “trending” list of each platform

8 / 20

slide-22
SLIDE 22

Motivation Background Methodology Analysis Conclusions

Methodology

Measurement Process

  • 1. Acquire Web page URLs for n videos from “trending” list of each platform
  • 2. Navigate to Web pages, determine source URLs of videos

◮ Video resolution: 480p ◮ Different source URL determination based on video service 8 / 20

slide-23
SLIDE 23

Motivation Background Methodology Analysis Conclusions

Methodology

Measurement Process

  • 1. Acquire Web page URLs for n videos from “trending” list of each platform
  • 2. Navigate to Web pages, determine source URLs of videos

◮ Video resolution: 480p ◮ Different source URL determination based on video service

  • 3. For each video:

3.1 connect() to determined media server (i.e., YouTube media server or DTube IPFS gateway) 3.2 Pass source URL to ExoPlayer for streaming/playout 3.3 Play video for one minute

8 / 20

slide-24
SLIDE 24

Motivation Background Methodology Analysis Conclusions

Methodology

Measurement Process

  • 1. Acquire Web page URLs for n videos from “trending” list of each platform
  • 2. Navigate to Web pages, determine source URLs of videos

◮ Video resolution: 480p ◮ Different source URL determination based on video service

  • 3. For each video:

3.1 connect() to determined media server (i.e., YouTube media server or DTube IPFS gateway) 3.2 Pass source URL to ExoPlayer for streaming/playout 3.3 Play video for one minute

  • 4. ICMP traceroute measurements to determined media servers
  • 5. Save and upload all measurements, schedule next iteration

8 / 20

slide-25
SLIDE 25

Motivation Background Methodology Analysis Conclusions

Methodology

Measurement Metrics

9 / 20

slide-26
SLIDE 26

Motivation Background Methodology Analysis Conclusions

Methodology

Measurement Metrics

◮ Total video duration ◮ TCP connect time to media server ◮ Startup delay ◮ traceroute: IP path length

9 / 20

slide-27
SLIDE 27

Motivation Background Methodology Analysis Conclusions

Methodology

Measurement Experiment

10 / 20

slide-28
SLIDE 28

Motivation Background Methodology Analysis Conclusions

Methodology

Measurement Experiment

◮ February 2019–November 2019 (10 months) ◮ >8,500 videos measured from both platforms combined ◮ Over both cellular (LTE) and WiFi (University network) ◮ Four mobile phones

◮ Three locations:

Munich (DE), Prague (CZ), San Diego (US)

◮ Four SIM card providers:

T-Mobile (DE), Vodafone (DE), o2 (DE), SIMPLE Mobile (US)

10 / 20

slide-29
SLIDE 29

Motivation Background Methodology Analysis Conclusions

Analysis

11 / 20

slide-30
SLIDE 30

Motivation Background Methodology Analysis Conclusions

Analysis

◮ How does decentralized video streaming compare with centralized services

in terms of performance?

◮ How distributed are such decentralized services? ◮ In which areas can decentralized video streaming be improved?

11 / 20

slide-31
SLIDE 31

Motivation Background Methodology Analysis Conclusions

Content Duration

250 500 750 1000 1250 1500 1750 2000 Content Duration [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

DTube YouTube

12 / 20

slide-32
SLIDE 32

Motivation Background Methodology Analysis Conclusions

Content Duration

250 500 750 1000 1250 1500 1750 2000 Content Duration [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

DTube YouTube

◮ YouTube: median 619 sec

◮ Longform videos (>10 minutes) allow additional advertisements 12 / 20

slide-33
SLIDE 33

Motivation Background Methodology Analysis Conclusions

Content Duration

250 500 750 1000 1250 1500 1750 2000 Content Duration [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

DTube YouTube

◮ YouTube: median 619 sec

◮ Longform videos (>10 minutes) allow additional advertisements

◮ DTube: median 323 sec

◮ Video length and monetization/incentive decoupled 12 / 20

slide-34
SLIDE 34

Motivation Background Methodology Analysis Conclusions

TCP Connect Time

50 100 150 200 250 300 350 400 TCP Connect Time [ms] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

50 100 150 200 250 300 350 TCP Connect Time [ms] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

T-Mobile (DE)

  • 2 (DE)

Vodafone (DE) SIMPLE Mobile (US) YouTube DTube

13 / 20

slide-35
SLIDE 35

Motivation Background Methodology Analysis Conclusions

TCP Connect Time

50 100 150 200 250 300 350 400 TCP Connect Time [ms] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

50 100 150 200 250 300 350 TCP Connect Time [ms] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

T-Mobile (DE)

  • 2 (DE)

Vodafone (DE) SIMPLE Mobile (US) YouTube DTube

◮ 75th percentiles by platform and network type:

◮ YouTube: 22 ms WiFi, 44 ms cellular ◮ DTube: 45 ms WiFi, 107 ms cellular 13 / 20

slide-36
SLIDE 36

Motivation Background Methodology Analysis Conclusions

TCP Connect Time

50 100 150 200 250 300 350 400 TCP Connect Time [ms] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

50 100 150 200 250 300 350 TCP Connect Time [ms] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

T-Mobile (DE)

  • 2 (DE)

Vodafone (DE) SIMPLE Mobile (US) YouTube DTube

◮ 75th percentiles by platform and network type:

◮ YouTube: 22 ms WiFi, 44 ms cellular ◮ DTube: 45 ms WiFi, 107 ms cellular

◮ 75th percentiles by cellular provider:

◮ 45–60 ms for both platforms and all ISPs, except SIMPLE to DTube (300 ms) 13 / 20

slide-37
SLIDE 37

Motivation Background Methodology Analysis Conclusions

TCP Connect Time

50 100 150 200 250 300 350 400 TCP Connect Time [ms] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

50 100 150 200 250 300 350 TCP Connect Time [ms] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

T-Mobile (DE)

  • 2 (DE)

Vodafone (DE) SIMPLE Mobile (US) YouTube DTube

◮ 75th percentiles by platform and network type:

◮ YouTube: 22 ms WiFi, 44 ms cellular ◮ DTube: 45 ms WiFi, 107 ms cellular

◮ 75th percentiles by cellular provider:

◮ 45–60 ms for both platforms and all ISPs, except SIMPLE to DTube (300 ms)

TCP connections to YouTube established in about half the time compared to DTube, although roughly within same order of magnitude (< 100 ms).

13 / 20

slide-38
SLIDE 38

Motivation Background Methodology Analysis Conclusions

Startup Delay

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 Startup Delay [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

2 4 6 8 10 12 14 Startup Delay [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

T-Mobile (DE)

  • 2 (DE)

Vodafone (DE) SIMPLE Mobile (US) YouTube DTube

14 / 20

slide-39
SLIDE 39

Motivation Background Methodology Analysis Conclusions

Startup Delay

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 Startup Delay [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

2 4 6 8 10 12 14 Startup Delay [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

T-Mobile (DE)

  • 2 (DE)

Vodafone (DE) SIMPLE Mobile (US) YouTube DTube

◮ 75th percentiles by platform and network type:

◮ YouTube: 0.82 sec WiFi, 1.35 sec cellular ◮ DTube: 3.2 sec WiFi, 5.8 sec cellular 14 / 20

slide-40
SLIDE 40

Motivation Background Methodology Analysis Conclusions

Startup Delay

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 Startup Delay [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

2 4 6 8 10 12 14 Startup Delay [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

T-Mobile (DE)

  • 2 (DE)

Vodafone (DE) SIMPLE Mobile (US) YouTube DTube

◮ 75th percentiles by platform and network type:

◮ YouTube: 0.82 sec WiFi, 1.35 sec cellular ◮ DTube: 3.2 sec WiFi, 5.8 sec cellular

◮ 75th percentiles by cellular provider:

◮ YouTube: 1–1.8 sec for all providers ◮ DTube: 3.1–4.6 sec for all providers but SIMPLE (9.8 sec) 14 / 20

slide-41
SLIDE 41

Motivation Background Methodology Analysis Conclusions

Startup Delay

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 Startup Delay [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

2 4 6 8 10 12 14 Startup Delay [s] 0.0 0.2 0.4 0.6 0.8 1.0 CDF

T-Mobile (DE)

  • 2 (DE)

Vodafone (DE) SIMPLE Mobile (US) YouTube DTube

◮ 75th percentiles by platform and network type:

◮ YouTube: 0.82 sec WiFi, 1.35 sec cellular ◮ DTube: 3.2 sec WiFi, 5.8 sec cellular

◮ 75th percentiles by cellular provider:

◮ YouTube: 1–1.8 sec for all providers ◮ DTube: 3.1–4.6 sec for all providers but SIMPLE (9.8 sec)

Startup delay for DTube about four times higher compared to YouTube; cellular measurements from the US to DTube perform significantly worse.

14 / 20

slide-42
SLIDE 42

Motivation Background Methodology Analysis Conclusions

IP Path Lengths

5 7 9 11 13 15 17 19 21 IP Path Length 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

15 / 20

slide-43
SLIDE 43

Motivation Background Methodology Analysis Conclusions

IP Path Lengths

◮ traceroute: failure rate of 17.3%

  • ver cellular network

◮ Highly varying success rate based

  • n cellular ISP

5 7 9 11 13 15 17 19 21 IP Path Length 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

15 / 20

slide-44
SLIDE 44

Motivation Background Methodology Analysis Conclusions

IP Path Lengths

◮ traceroute: failure rate of 17.3%

  • ver cellular network

◮ Highly varying success rate based

  • n cellular ISP

5 7 9 11 13 15 17 19 21 IP Path Length 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

◮ By platform and network type:

◮ YouTube: reachable within 10 IP hops (WiFi: 93.9%, cellular: 86.0%) ◮ DTube: reachable within 10 IP hops only for 4.6% (WiFi);

minimum over cellular 11 IP hops (29.8%)

◮ Paths to YouTube shorter by 7–8 IP hops in comparison 15 / 20

slide-45
SLIDE 45

Motivation Background Methodology Analysis Conclusions

IP Path Lengths

◮ traceroute: failure rate of 17.3%

  • ver cellular network

◮ Highly varying success rate based

  • n cellular ISP

5 7 9 11 13 15 17 19 21 IP Path Length 0.0 0.2 0.4 0.6 0.8 1.0 CDF

YouTube (WiFi) YouTube (Cellular) DTube (WiFi) DTube (Cellular)

◮ By platform and network type:

◮ YouTube: reachable within 10 IP hops (WiFi: 93.9%, cellular: 86.0%) ◮ DTube: reachable within 10 IP hops only for 4.6% (WiFi);

minimum over cellular 11 IP hops (29.8%)

◮ Paths to YouTube shorter by 7–8 IP hops in comparison

traceroute success rates over cellular network highly depend on ISP. Around 90% of the YouTube destinations within 10 IP hops; for DTube, more than 95% of the destinations beyond 10 IP hops.

15 / 20

slide-46
SLIDE 46

Motivation Background Methodology Analysis Conclusions

Destination ASes

16 / 20

slide-47
SLIDE 47

Motivation Background Methodology Analysis Conclusions

Destination ASes

YouTube

Google LLC, US (AS15169) MNET-AS, DE (AS8767) O2, CZ (AS5610) T-Mobile, US (AS21928) TDDE-ASN1, DE (AS6805) All cellular 921 (18.9%) 0 (0%) 0 (0%) 105 (2.2%) 35 (0.7%) 1061 (21.7%) WiFi 3623 (74.2%) 3 (0.1%) 196 (4.0%) 0 (0%) 0 (0%) 3822 (78.3%) All 4544 (93.1%) 3 (0.1%) 196 (4.0%) 105 (2.2%) 35 (0.7%) 4883 (100%)

◮ YouTube: Videos streamed from nearby ISP ASes (caches) and Google AS

16 / 20

slide-48
SLIDE 48

Motivation Background Methodology Analysis Conclusions

Destination ASes

YouTube

Google LLC, US (AS15169) MNET-AS, DE (AS8767) O2, CZ (AS5610) T-Mobile, US (AS21928) TDDE-ASN1, DE (AS6805) All cellular 921 (18.9%) 0 (0%) 0 (0%) 105 (2.2%) 35 (0.7%) 1061 (21.7%) WiFi 3623 (74.2%) 3 (0.1%) 196 (4.0%) 0 (0%) 0 (0%) 3822 (78.3%) All 4544 (93.1%) 3 (0.1%) 196 (4.0%) 105 (2.2%) 35 (0.7%) 4883 (100%)

DTube

OVH, FR (AS16276) All 634 (28.9%) 634 (28.9%) 1556 (71.1%) 1556 (71.1%) 2190 (100%) 2190 (100%)

◮ YouTube: Videos streamed from nearby ISP ASes (caches) and Google AS ◮ DTube: All videos streamed from OVH AS16276

16 / 20

slide-49
SLIDE 49

Motivation Background Methodology Analysis Conclusions

Destination ASes

YouTube

Google LLC, US (AS15169) MNET-AS, DE (AS8767) O2, CZ (AS5610) T-Mobile, US (AS21928) TDDE-ASN1, DE (AS6805) All cellular 921 (18.9%) 0 (0%) 0 (0%) 105 (2.2%) 35 (0.7%) 1061 (21.7%) WiFi 3623 (74.2%) 3 (0.1%) 196 (4.0%) 0 (0%) 0 (0%) 3822 (78.3%) All 4544 (93.1%) 3 (0.1%) 196 (4.0%) 105 (2.2%) 35 (0.7%) 4883 (100%)

DTube

OVH, FR (AS16276) All 634 (28.9%) 634 (28.9%) 1556 (71.1%) 1556 (71.1%) 2190 (100%) 2190 (100%)

◮ YouTube: Videos streamed from nearby ISP ASes (caches) and Google AS ◮ DTube: All videos streamed from OVH AS16276

All traces to DTube end in OVH AS (FR), while YouTube traces end in ISP caches and Google ASes, indicating locational centralization for DTube.

16 / 20

slide-50
SLIDE 50

Motivation Background Methodology Analysis Conclusions

Conclusions

17 / 20

slide-51
SLIDE 51

Motivation Background Methodology Analysis Conclusions

Conclusions: Limitations

18 / 20

slide-52
SLIDE 52

Motivation Background Methodology Analysis Conclusions

Conclusions: Limitations

◮ Limited set of (network-related) measurement metrics ◮ Limited number of measurement configurations, geographical bias ◮ DTube as only representative for decentralized video streaming ◮ Several additions/changes to DTube (video sources, tokens, ...)

18 / 20

slide-53
SLIDE 53

Motivation Background Methodology Analysis Conclusions

Conclusions: Summary

19 / 20

slide-54
SLIDE 54

Motivation Background Methodology Analysis Conclusions

Conclusions: Summary

◮ Higher connection and startup delays for DTube compared to YouTube

◮ TCP connect times about twice as high (WiFi 45 ms, cellular 107 ms) ◮ Startup delay about four times higher (WiFi 3.2 sec, cellular 5.8 sec) ◮ IP path lengths higher by 7–8 IP hops 19 / 20

slide-55
SLIDE 55

Motivation Background Methodology Analysis Conclusions

Conclusions: Summary

◮ Higher connection and startup delays for DTube compared to YouTube

◮ TCP connect times about twice as high (WiFi 45 ms, cellular 107 ms) ◮ Startup delay about four times higher (WiFi 3.2 sec, cellular 5.8 sec) ◮ IP path lengths higher by 7–8 IP hops

◮ Locational centralization of DTube

◮ Private IPFS network, lack of distributed content servers ◮ Low number of videos from public IPFS network/gateways 19 / 20

slide-56
SLIDE 56

Motivation Background Methodology Analysis Conclusions

Conclusions: Summary

◮ Higher connection and startup delays for DTube compared to YouTube

◮ TCP connect times about twice as high (WiFi 45 ms, cellular 107 ms) ◮ Startup delay about four times higher (WiFi 3.2 sec, cellular 5.8 sec) ◮ IP path lengths higher by 7–8 IP hops

◮ Locational centralization of DTube

◮ Private IPFS network, lack of distributed content servers ◮ Low number of videos from public IPFS network/gateways

◮ However, DTube and decentralized technologies still under development

19 / 20

slide-57
SLIDE 57

Motivation Background Methodology Analysis Conclusions

Conclusions: Summary

◮ Higher connection and startup delays for DTube compared to YouTube

◮ TCP connect times about twice as high (WiFi 45 ms, cellular 107 ms) ◮ Startup delay about four times higher (WiFi 3.2 sec, cellular 5.8 sec) ◮ IP path lengths higher by 7–8 IP hops

◮ Locational centralization of DTube

◮ Private IPFS network, lack of distributed content servers ◮ Low number of videos from public IPFS network/gateways

◮ However, DTube and decentralized technologies still under development ◮ Open-source Android app: extensible for other streaming services

19 / 20

slide-58
SLIDE 58

Motivation Background Methodology Analysis Conclusions

Conclusions: Summary

◮ Higher connection and startup delays for DTube compared to YouTube

◮ TCP connect times about twice as high (WiFi 45 ms, cellular 107 ms) ◮ Startup delay about four times higher (WiFi 3.2 sec, cellular 5.8 sec) ◮ IP path lengths higher by 7–8 IP hops

◮ Locational centralization of DTube

◮ Private IPFS network, lack of distributed content servers ◮ Low number of videos from public IPFS network/gateways

◮ However, DTube and decentralized technologies still under development ◮ Open-source Android app: extensible for other streaming services

https://github.com/tv-doan/ifip-net-2020-app (source code) https://github.com/tv-doan/ifip-net-2020-analysis (artifacts)

App Analysis

[ trinhviet.doan | dat.pham | markus.oberprieler | vaibhav.bajpai ]@tum.de

19 / 20

slide-59
SLIDE 59

Motivation Background Methodology Analysis Conclusions

References

[1] Internet Society, “Internet Society Global Internet Report: Consolidation in the Internet Economy,” 2019. https://future.internetsociety.org/2019/. [2] J. Arkko, B. Trammell, M. Nottingham, C. Huitema, M. Thomson, J. Tantsura, and

  • N. ten Oever, “Considerations on Internet Consolidation and the Internet

Architecture,” 2019. https://www.ietf.org/archive/id/ draft-arkko-iab-internet-consolidation-02.txt. [3] Journal of Cyber Policy, “Special Issue: Consolidation of the Internet,” 2020. https://www.tandfonline.com/toc/rcyb20/5/1. [4] Sandvine, “Global Internet Phenomena Report,” 2019. https://bit.ly/3cvN5Qi.

20 / 20