Measuring the Efgects of Happy Eyeballs Takeway Limitations - - PowerPoint PPT Presentation

▶
measuring the efgects of happy eyeballs
SMART_READER_LITE
LIVE PREVIEW

Measuring the Efgects of Happy Eyeballs Takeway Limitations - - PowerPoint PPT Presentation

Jan 31, 2024 166 likes •428 views

Introduction Motivation Leone Project: leone-project.eu Supported by: October 2016 Jacobs University, Bremen Jrgen Schnwlder Joint work with Applied Networking Research Workshop Published at: Madrid, Spain RIPE 73 Jacobs

slide-1
SLIDE 1

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Measuring the Efgects of Happy Eyeballs

Vaibhav Bajpai

Jacobs University, Bremen

RIPE 73 Madrid, Spain

Published at: Applied Networking Research Workshop July 2016: https://goo.gl/ZiIcQV

Joint work with Jürgen Schönwälder Jacobs University, Bremen October 2016

Supported by: Flamingo Project: fmamingo-project.eu Leone Project: leone-project.eu 1 / 18

slide-2
SLIDE 2

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Introduction | Motivation

▶ HE timer (300 ms) was chosen (2012) when broken IPv6 connectivity was prevalent.

▶ Largely attributed to failures caused by Teredo [1] and 6to4 relays [2]. ▶ Even in situations where relays work, Teredo / 6to4 add noticeable latency [3, 4].

▶ Tiese transition mechanisms have declined over the years due to efgorts such as −

2013 Microsofu stopped Teredo on Windows and deactivated public Teredo servers [5]. 2015 Tie 6to4 anycast prefjx has been obsoleted [6].

▶ Consequentely, failure rates over IPv6 [7] have dropped signifjcantly −

Overall Native 2011 40% 5.3% 2015 3.5% 2%

2 / 18

slide-3
SLIDE 3

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Introduction | Browser Implementations

Fragmentation of HE is visible in browser implementations today − 2011 Chrome uses 300 ms [8].

[since v11]

2011 Safari uses history of witnessed latencies [9].

[since OS X 10.7]

2012 Opera uses parallel TCP connections [10].

[since v12.10]

2012 Firefox uses parallel TCP connections [11].

[since v15]

Firefox [network.http.fast-fallback-to-IPv4=false] uses 250 ms. 2015 Safari uses 25 ms + history of witnessed latencies [12].

[since OS X 10.11 / iOS 9]

Tiese HE timer values are arbitrarily chosen. What is the right timer value?

3 / 18

slide-4
SLIDE 4

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Introduction | Research Contributions

We measure against ALEXA top 10K websites for 3 years (2013 - 2016)

  • 1. TCP connect times to websites over IPv6 have considerably improved over time.
  • 2. 18% of websites are faster over IPv6 with 91% being at most 1 ms slower (May ’16).
  • 3. HE (300 ms) makes 99% of websites prefer IPv6 more than 98% of the time.
  • 4. Slower IPv6 connections are preferred in ∼90% of the cases.
  • 5. Lowering HE (150 ms) gives a margin benefjt of 10% and retains same preference levels.

4 / 18

slide-5
SLIDE 5

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Methodology1

1Please see previous work [13] for a more detailed description of our methodology 5 / 18

slide-6
SLIDE 6

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Methodology | Metrics and Implementation

▶ Uses getaddrinfo(…) to resolve service names. ▶ Uses non-blocking TCP connect(…) calls. ▶ DNS resolution time is not accounted. ▶ Can read multiple service names as arguments. ▶ Can read service names list from a fjle. ▶ File locking capability. ▶ Sets a delay between connect(…) ; avoids SYN fmoods. ▶ Can produce both human-readable & CSV output. ▶ Cross-compiled for OpenWrt; Running on SamKnows.

happy 1) endpoint 2) endpoint 3) endpoint ... n) endpoint connection establishment times (µs) 1) service name 2) port

happy.vaibhavbajpai.com.

% happy -q 1 -m www.google.com www.facebook.com HAPPY.0;1360681039;OK;www.google.com;80;173.194.69.105;8626 HAPPY.0;1360681039;OK;www.google.com;80;2a00:1450:4008:c01::69;8884

6 / 18

slide-7
SLIDE 7

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Methodology | Selection of Websites

▶ We use the ALEXA top 10K websites as

measurement targets [13].

  • 1. www.google.com
  • 2. www.facebook.com
  • 3. www.youtube.com
  • 4. www.yahoo.com
  • 5. www.wikipedia.org
  • 6. www.qq.com
  • 7. www.blogspot.com
  • 8. …

7 / 18

slide-8
SLIDE 8

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Methodology | Measurement Setup

Tie happy test repeats every hour.

DSL/Cable Modem SamKnows Tests Probe

ALEXA Top 10K Websites results HTTPS POST TCP connect(...)

IPv6 IPv4

happy Data Collector

8 / 18

slide-9
SLIDE 9

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Methodology | Measurement Trial

NETWORK TYPE # RESIDENTIAL 55 NREN / RESEARCH 11 BUSINESS / DATACENTER 09 OPERATOR LAB 04 IXP 01 RIR # RIPE 42 ARIN 29 APNIC 07 AFRINIC 01 LACNIC 01

We measure from 80 dual-stacked SamKnows [14] probes.

9 / 18

slide-10
SLIDE 10

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Data Analysis

[2013 - 2016]

10 / 18

slide-11
SLIDE 11

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Data Analysis | Trends (2013 - 2016)

∆sa(u) = t4(u) − t6(u)

where t(u) is the time taken to establish TCP connection to website u.

−150 −100 −50 50 www.bing.com www.facebook.com www.wikipedia.org www.youtube.com 2013 2014 2015 2016 02 05 08 11 02 05 08 11 02 05 08 11 02 05 −60 −40 −20 20 ∆sa (ms) www.blogspot.* www.google.* www.netflix.com www.yahoo.com

▶ TCP connect times to popular websites over IPv6 have considerably improved over time.

11 / 18

slide-12
SLIDE 12

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Data Analysis | Who connects faster?

ALEXA top 10K websites (as of May 2016):

▶ 18% are faster over IPv6. ▶ 91% of the rest are at most 1 ms slower. ▶ 3% are at least 10 ms slower. ▶ 1% are at least 100 ms slower.

−1.0 −0.5 0.0 0.5 1.0 ∆sa (ms) 0.0 0.2 0.4 0.6 0.8 1.0 CDF netflix yahoo google youtube linkedin microsoft facebook wikipedia ALEXA (10K) [05/2016] ∆sa(u) = t4(u) − t6(u) 12 / 18

slide-13
SLIDE 13

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Data Analysis | Preference

▶ Only ∼1% of samples above HE

timer value > 300 ms

▶ A 300 ms HE timer value leaves

2% chance for IPv4.

▶ 99% of top 10K ALEXA prefer

IPv6 98% of time.

10-1 100 101 102 103 104 TCP Connect Times (ms) 0.0 0.2 0.4 0.6 0.8 1.0

CDF

300 ms IPv6 (189K) IPv4 (189K) ['13 - '16] 96% 97% 98% 99% 100% 0.0 0.2 0.4 0.6 0.8 1.0 CCDF PROBES (80) ALEXA (10K) ['13 - '16] Preference (300 ms) 13 / 18

slide-14
SLIDE 14

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Data Analysis | Slowness

Samples where HE prefers IPv6 −

▶ HE prefers slower IPv6

connections 90% of the time.

▶ Absolute difgerence is not that far

apart from IPv4

▶ 30% − at least 1 ms slower. ▶ 7% − at least 10 ms slower. −40 −30 −20 −10 10 ∆sa (ms) 0.0 0.2 0.4 0.6 0.8 1.0 CDF 1% 2% 7% 30% 93% 99% 189K −0.6 −0.4 −0.2 0.0 0.2 ∆sr 0.0 0.2 0.4 0.6 0.8 1.0 CDF 97% 90% 57% 8% 2% 189K ['13 - '16] ∆sa(u) = t4(u) − t6(u) ∆sr(u) = t4(u)−t6(u)

t4(u)

Can a lower HE timer provide same preference over IPv6 but not penalise IPv4 when it’s faster?

14 / 18

slide-15
SLIDE 15

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Data Analysis | Lowering HE Timer

Are we ready to disable HE entirely?

2013 2014 2015 2016 02 05 08 11 02 05 08 11 02 05 08 11 02 05 0% 20% 40% 60% 80% 100% IPv6 Preference 0 ms 150 ms 300 ms

▶ 18% of ALEXA top 10K websites are faster (see slide 17) over IPv6 today. ▶ Parallel TCP connections2 (HE with 0 ms timer) will hamper IPv6 preference. ▶ HE timer today still should give IPv6 a fair chance to succeed.

2such as used by Firefox and Opera today 15 / 18

slide-16
SLIDE 16

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Data Analysis | Lowering HE Timer

▶ We control two3 parameters and

lower the HE timer value.

▶ Each data point is the 1th

percentile preference towards ALEXA 10K websites.

▶ Lowering to 150 ms retains

preference levels over IPv6.

▶ We get margin benefjt of 10%

(18.9K) because timer cuts early.

50 100 150 200 250 300 HE timer (ms) 0% 20% 40% 60% 80% 100% Preference 150 ms ALEXA (10K) ['13 - '16] −0.6 −0.4 −0.2 0.0 0.2 ∆sr 0.0 0.2 0.4 0.6 0.8 1.0 CDF 80% 189K ['13 - '16] Slowness (150ms) 399% ALEXA top 10K websites prefer IPv6 connections 98.6% of the time 16 / 18

slide-17
SLIDE 17

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Limitations

  • 1. Tie comparison refmects the performance as seen over TCP port 80 only.
  • 2. Tie measurements cover ALEXA top 10K websites only.
  • 3. Tie results are biased by our vantage points (centered largely around EU, US and JP).

17 / 18

slide-18
SLIDE 18

Introduction

Motivation Browser Implementations Research Contributions

Methodology

Metrics and Implementation Selection of Websites Measurement Setup Measurement Trial

Data Analysis

Trends Who connects faster? Preference Slowness Lowering HE Timer

Limitations Takeway

Takeway

  • 1. TCP connect times to websites over IPv6 have considerably improved over time.
  • 2. 18% of websites are faster over IPv6 with 91% being at most 1 ms slower (May ’16).
  • 3. HE (300 ms) makes 99% of websites prefer IPv6 more than 98% of the time.
  • 4. Slower IPv6 connections are preferred in ∼90% of the cases.
  • 5. Lowering HE (150 ms) gives a margin benefjt of 10% and retains same preference levels.

www.vaibhavbajpai.com v.bajpai@jacobs-university.de | @bajpaivaibhav

18 / 18

slide-19
SLIDE 19

Background

Research Question

Related Work

Appendix

1 / 6

slide-20
SLIDE 20

Background

Research Question

Related Work

References

[1]

  • C. Huitema, “Teredo: Tunneling IPv6 over UDP through Network

NATs,” RFC 4380, Internet Engineering Task Force, Feb. 2006, https://tools.ietf.org/html/rfc4380. [2]

  • B. Carpenter and K. Moore, “Connection of IPv6 Domains via IPv4

Clouds,” RFC 3056, Internet Engineering Task Force, Feb. 2001, https://tools.ietf.org/html/rfc3056. [3]

  • S. Zander, L. L. H. Andrew, G. J. Armitage, G. Huston, and
  • G. Michaelson, “Investigating the IPv6 Teredo Tunnelling Capability

and Performance of Internet Clients,” ser. Computer Communication Review (CCR) ’12, 2012, pp. 13–20. [Online]. Available: http://doi.acm.org/10.1145/2378956.2378959 [4]

  • L. Colitti, S. H. Gunderson, E. Kline, and T. Refjce, “Evaluating IPv6

Adoption in the Internet,” ser. Passive and Active Measurement Conference (PAM) ’10, 2010, pp. 141–150. [Online]. Available: http://dx.doi.org/10.1007/978-3-642-12334-4_15 [5] “Christopher Palmer - Teredo at Microsofu: Present and Future,” http://goo.gl/9I65Wy, [Online; accessed 10-February-2016]. [6]

  • O. Troan and B. Carpenter, “Deprecating the Anycast Prefjx for 6to4

Relay Routers,” RFC 7526, Internet Engineering Task Force, May 2015, https://tools.ietf.org/html/rfc7526. [7] “Geofg Huston - Measuring IPv6 Performance,” https://goo.gl/n78W1t, [Online; accessed 10-February-2016]. [8] “Google Chrome - Revision 85934: Add a fallback socket connect() for IPv6.” https://goo.gl/nPhilZ, [Online; accessed 25-January-2016]. [9]

  • J. Graessley, “Apple - Lion and IPv6,” http://goo.gl/uAPlV8, [Online;

accessed 25-January-2016]. [10] “Opera 12.10 - Changelog,” http://goo.gl/MGsn4K, [Online; accessed 25-Jan-2016]. [11] “Mozilla Firefox 15 - Release Notes,” http://goo.gl/hA15eu, [Online; accessed 25-January-2016]. [12]

  • D. Schinazi, “Apple and IPv6 - Happy Eyeballs,”

https://goo.gl/1nzMs6, [Online; accessed 25-January-2016]. [13]

  • V. Bajpai and J. Schönwälder, “IPv4 versus IPv6 - who connects

faster?” ser. IFIP NETWORKING ’15, 2015, pp. 1–9. [Online]. Available: http://dx.doi.org/10.1109/IFIPNetworking.2015.7145323 [14] ——, “A Survey on Internet Performance Measurement Platforms and Related Standardization Efgorts,” ser. IEEE Communications Surveys and Tutorials (COMST) ’15, 2015, pp. 1313–1341. [Online]. Available: http://dx.doi.org/10.1109/COMST.2015.2418435 [15]

  • D. Tialer, R. Draves, A. Matsumoto, and T. Chown, “Default

Address Selection for Internet Protocol Version 6 (IPv6),” RFC 6724 (Proposed Standard), Internet Engineering Task Force, Sep. 2012. [Online]. Available: http://www.ietf.org/rfc/rfc6724.txt 1 / 6

slide-21
SLIDE 21

Background

Research Question

Related Work [16] “Teemu Savolainen - Experiences of host behavior in broken IPv6 networks,” http://goo.gl/4NnRiH, [Online; accessed 25-January-2016]. [17]

  • P. Richter, M. Allman, R. Bush, and V. Paxson, “A Primer on IPv4

Scarcity,” ser. Computer Communication Review (CCR), vol. 45,

  • no. 2.

New York, NY, USA: ACM, Apr. 2015, pp. 21–31. [Online]. Available: http://doi.acm.org/10.1145/2766330.2766335 [18] “Internet Society - World IPv6 Launch,” http://www.worldipv6launch.org, [Online; accessed 11-January-2016]. [19] “Google - IPv6 Adoption Statistics,” http://www.google.com/intl/en/ipv6/statistics.html, [Online; accessed 11-January-2016]. [20] “Lorenzo Colitti - Google no longer returning AAAA records?” https://goo.gl/6Z7gZM, [Online; accessed 11-January-2016]. [21] “Emile Aben - Hampering Eyeballs: Observations on Two Happy Eyeballs Implementations,” https://goo.gl/3xVUIO, [Online; accessed 10-February-2016]. [22] “Geofg Huston - Dual Stack Esotropia,” http://goo.gl/N1qUib, [Online; accessed 10-February-2016]. [23] “Geofg Huston - Bemused Eyeballs: Tailoring Dual Stack Applications for a CGN Environment,” http://goo.gl/LMPc4h, [Online; accessed 10-February-2016]. [24]

  • F. Baker, “Testing Eyeball Happiness,” RFC 6556, Internet

Engineering Task Force, 2012, https://tools.ietf.org/html/rfc6556. [25]

  • S. Zander, L. L. H. Andrew, G. J. Armitage, G. Huston, and
  • G. Michaelson, “Mitigating Sampling Error when Measuring Internet

Client IPv6 Capabilities,” ser. Internet Measurement Conference (IMC) ’12, 2012, pp. 87–100. [Online]. Available: http://doi.acm.org/10.1145/2398776.2398787 [26]

  • V. Bajpai and J. Schönwälder, “Measuring the Efgects of Happy

Eyeballs,” Internet Engineering Task Force, Internet-Drafu drafu-bajpai-happy-01, Jul. 2013, work in Progress. [Online]. Available: http://tools.ietf.org/html/draft-bajpai-happy-01 [27]

  • S. Ahsan, V. Bajpai, J. Ott, and J. Schönwälder, “Measuring YouTube

from Dual-Stacked Hosts,” ser. Passive and Active Measurement Conference (PAM) ’15, 2015, pp. 249–261. [Online]. Available: http://dx.doi.org/10.1007/978-3-319-15509-8_19 2 / 6

slide-22
SLIDE 22

Background

Research Question

Related Work

Introduction | getaddrinfo(…) behavior

1) native IPv6 routes ... 2) native IPv4 routes ... 3) IPv4-IPv6 Transitioning routes getaddrinfo(...) preference: TCP connection request

▶ returns a list of endpoints in an order that prioritizes an IPv6-upgrade path. ▶ Tie order is prescribed by RFC 6724 [15] and /etc/gai.conf ▶ Iterating sequentially over the list of IP endpoints has repercussions −

▶ Broken IPv6 connectivity makes apps stall for several seconds before trying IPv4. ▶ Studies have reported [16] browser connection timeouts in the order of 20

seconds.

2 / 6

slide-23
SLIDE 23

Background

Research Question

Related Work

Introduction | Happy Eyeballs [RFC 6555]

HE helps prevent bad QoE in situations where IPv6 connectivity is broken.

t0 t0 + 300ms

time IPv6 IPv4 Happy Eyeballs [RFC 6555]

Design Goals −

▶ Honor the destination address selection policy [RFC 6724] [15]. ▶ Quickly fallback to IPv4 when IPv6 connectivity is broken. ▶ Give a fair chance for IPv6 to succeed.

3 / 6

slide-24
SLIDE 24

Background

Research Question

Related Work

Introduction | Motivation

IPv6 landscape has changed today −

▶ 4/5 RIRs have exhausted available pool of IPv4 address space [17].

APNIC Apr′11 RIPE Sep′12 LACNIC Jun′14 ARIN Sep′15

▶ Large IPv6 broadband rollouts4 since World IPv6 Launch Day in 2012 [18]. ▶ IPv6 global adoption at ∼12.2% (native) with Teredo / 6to4 at ∼0.01% [19] (July 2016) ▶ Google over IPv6 (whitelist) program replaced by a Google IPv6 blacklist [13]. ▶ Google will not return AAAA to resolvers where latency over IPv6 > 100 ms worse [20].

4Comcast, Deutsche Telekom AG, AT&T, Verizon Wireless, T-Mobile USA 4 / 6

slide-25
SLIDE 25

Background

Research Question

Related Work

Introduction | Research Questions

Tie efgects of HE (300 ms) on the QoE of a dual-stacked user remains largely unclear. We want to know −

▶ In what percentage of cases HE makes a bad decision of choosing IPv6 when it’s slower? ▶ In such situations what is the amount of imposition (in terms of latency impact) a

dual-stacked user has to pay as a result of the high HE timer (300 ms) value? Applications apply HE not only where IPv6 is broken, but also when IPv6 is comparable.

5 / 6

slide-26
SLIDE 26

Background

Research Question

Related Work

Related Work

2011 - 2012 Studies [21, 22, 23] have analyzed HE implementations.

▶ Chrome reduces degraded user experience when IPv6 is broken. ▶ Firefox [network.http.fast-fallback-to-IPv4=false] behaves similar to Chrome. ▶ Safari prefers IPv4 even when IPv6 connectivity is similar (hampering eyeballs).

Tiese studies are dated. HE implementations have changed with time (see slide 7). 2012 Baker [24] describes HE metrics and testbed confjgurations. 2012 Zander [25] showed that 75% of the connection attempts preferred5 IPv6. 2013 We [26] showed that HE never prefers IPv6 using Teredo. 2015 We [27] showed that HE prefers YouTube over IPv6 even when IPv4 performs better.

5In this work, we show that this preference has increased to 98% today 6 / 6