SlideShare a Scribd company logo
Measuring the Internet Economy:
How Networks Create Value
Bill Woodcoc
k

Executive Directo
r

Packet Clearing House
Version 1.
0

bdNOG1
3

June 10-12, 2021
Why Measure the Internet?
Tactical reasons:
To understand where the value of the product you’re selling comes from

As input to calculating Average Per-Bit Delivery Costs (APBDC)

So that you can bill your customers in di erent and more useful ways

To optimize and plan network interconnection (peering and transit)

Strategic reasons:
So you’ll know your business when you talk with your regulator or customers

To justify infrastructural investment and attention from policy-makers
Why Measure the Internet?
Although economists argue about methodologies and speci cs, there’s general
agreement that the Internet is responsible for a quarter to a third of new productivity in
most OECD countries, and a double-digit percentage globally.

It’s the responsibility of governmental policy-makers, of regulators, and of people in the
involved industries, to understand and support such engines of growth.

The Internet allows people without work to
fi
nd jobs, children to continue their education
during a pandemic, businesses to receive orders and ful ll them. The economic activity
produces tax revenue which pays for schools, roads, law enforcement, and healthcare.

Regardless of whether individual customers appreciate the work that you do, society
and posterity depend upon your work and its success.
Theodore Vail used the Network E
ff
ect to justify an AT&T monopoly in 1908.

Value increases as a function of the number of users of a compatible
product or service.

In communications, value increases as a function of the number of other
people who can be reached through a network.

The Network Effect
The Network Effect
Users: 2

Value: 1

Value per user: 0.5

Marginal value

of next user: 2
Users: 4

Value: 6

Value per user: 1.5

Marginal value

of next user: 4
Users: 6

Value: 15

Value per user: 2.5

Marginal value

of next user: 6
Users: 5

Value: 10

Value per user: 2

Marginal value

of next user: 5
Users: 3

Value: 3

Value per user: 1

Marginal value

of next user: 3
The Network Effect
Source: CAIDA
The Network Effect
Network e
ff
ects apply not only to underlying communications infrastructure like ber
cables, but at all layers of the network. Some examples:

Layer 2: Ethernet adjacencies

Layer 3: Global reachability with the Internet Protocol

Layer 5: BGP route-servers like those which allowed the IXPs in Jakarta and Sao
Paulo to so quickly outpace those of Europe and the United States

Layer 8: Speakers of a common language, or people who share a common
understanding of a scienti
fi
c principle
Internet Service Provider
Lifecycle Phase 1:


Simple Aggregator
Single Transit Provider
Customers
ISP Network
IXPs
Internet Service Provider
Lifecycle Phase 1:


Simple Aggregator
Single Transit Provider
Customers
ISP Network
IXPs
Vertical connections are
transit: service travels
downward, money travels
upward.
Redundant Transit Providers
Customers
ISP Network
IXPs
Internet Service Provider
Lifecycle Phase 2:


Redundancy & Efficiency
Redundant Transit Providers
Customers
ISP Network Single IXP
IXPs
Internet Service Provider
Lifecycle Phase 3:


Local Peering
Redundant Transit Providers
Customers
ISP Network Single IXP
IXPs
Horizontal connections are peering:
bandwidth is exchanged, but money
is not.
Internet Service Provider
Lifecycle Phase 3:


Local Peering
Redundant Transit Providers
Customers
ISP Network
Multipl
e

IXPs
IXPs
Internet Service Provider
Lifecycle Phase 4:


Diverse Peering
Measuring the Internet Economy: How Networks Create Value
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
C C
C C
C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
C
C
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2
1
1 1
2
C
C
1
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
3
2
2
2
1
1 1
3
2
4
C
C
1
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
IXP IXP IXP
IXP
4
4
3
2
2
2
1
1 1
3
2
IXP
4
C
C
1
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
IXP IXP IXP
IXP
4
3
2
2
2
1
1 1
3
2
4
IXP
4
C
C
1
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
IXP IXP IXP
IXP
4
3
2
2
2
1
1 1
2
3
4
IXP
4
C
C
1
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
IXP IXP IXP
IXP
4
3
2
2
2
1
1 1
3
4
IXP
4
C
C
1
2
C C
C C
C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
C C
C C
C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
Transit Cone
Customers: 3

Value: 6
C C
C C
C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
Transit Cone
Customers: 5

Value: 15
C C
C C
C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
Transit Cone
Customers: 9

Value: 45
C C
C C
C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
Transit Cone
Customers: 16

Value: 136
C C
C C
C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
Customers: 24

Value: 300
Transit Cone + Peering
Source: CAIDA
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
C C
C C
C
C
C C
C C C
C
C
C
C C
C
C
C
C
C
C
C
C
IXP
IXP IXP IXP
IXP
4
4
4
3 3
2
2
2 2
1
1
1
1
IXP
IXP IXP
C
C
C
C
C C
So What Do We Measure?
The vast majority of bandwidth is produced in Internet exchange points.

Bandwidth produced in IXPs comes from interconnections between participating
networks across the switch fabric, or across passive ber crossconnects.

Aggregate bandwidth produced on the switch fabric is typically a public number.

The amount of bandwidth produced on private pieces of ber is known only to
the two parties to each interconnection, but traceroutes will reveal whether the
connection is occurring across the switch fabric or a private connection.

A tiny remaining fraction of bandwidth is produced within transit cones, and
traceroute analysis can identify this, but not quantify it, as well.
So What Do We Measure?
The ratio of bandwidth produced through each of these three mechanisms is
not directly measurable, for many reasons.

The ratio of interconnections facilitated by each of the three mechanisms can
be determined by analyzing traceroutes, but each of the three types have
characteristically di
ff
erent tra
ffi
c capacities and average utilization. 

Di
ff
erent business practices and regulatory environments strongly a ect the
ratios of tra
ffi
c generated using each mechanism:

• Regulatorily-protected monopoly incumbents in many developing economies

• Data-collection-driven preference for switching over crossconnects
So What Do We Measure?
More speci
fi
cally, at exchange points, we sum the count of outbound bits
across all ISP-facing switch ports.

Not inbound, because that wouldn’t catch multicast, but would catch
bogons and subsequently-mitigated DDoS.

Not all ports, because that would double-count inter-switch tra c.

If we want to get really fancy, we can deduplicate broadcast tra c, but
there really shouldn’t be broadcast tra
ffi
c on an IX subnet.
So What Do We Measure?
In an ISP network, measurement is used for both engineering and billing.
We typically use
fl
ow-based measurement, in addition to summing bits in
both directions at the peering and upstream transit edges.

This allows a “local tra
ffi
c is free, international tra c is metered” two-
category billing model.

If you capture customer-facing port bit-counts as well, that enables a “on-
net, peer, and transit” three-category model.

Old-style customer-port-bit-count-only doesn’t allow di erentiated
pricing, nor does it tell you which customers are costing you most.
What Questions Can We Answer?
If we measure Internet tra
ffi
c comprehensively, ideally using both bit-counts
and sparsely-sampled
fl
ows, we can answer many questions:

• Which of our customers make us pro
fi
table, and which cost us money?

• Is our country or region a net importer or net exporter of bandwidth? 

• Where is the bandwidth our users are consuming produced?

• Where are the consumers of the bandwidth we’re producing?

• Which protocols and applications are gaining momentum, and which are
falling into disuse?
Thanks, and Questions?
Bill Woodcoc
k

Executive Directo
r

Packet Clearing Hous
e

woody@pch.ne
t

https://pch.net

More Related Content

PDF
Broadcast storms in service provider network, Nafeez Islam
PDF
Lifting the Lid on Lawful Intercept
PDF
Social CDN Business : Paid or Free - bdNOG12 Panel Discussion
PDF
Challenges for BdREN in COVID Environment
PDF
Learning from failures
PDF
12 (IDNOG02) SGIX and Singapore Internet Landscape by Kerk Chun Sing
PDF
13 (IDNOG02) Bridging ASEAN 2.0 by Zulhaidi Mokti
PPTX
Broad Sky SD-WAN September 2018
Broadcast storms in service provider network, Nafeez Islam
Lifting the Lid on Lawful Intercept
Social CDN Business : Paid or Free - bdNOG12 Panel Discussion
Challenges for BdREN in COVID Environment
Learning from failures
12 (IDNOG02) SGIX and Singapore Internet Landscape by Kerk Chun Sing
13 (IDNOG02) Bridging ASEAN 2.0 by Zulhaidi Mokti
Broad Sky SD-WAN September 2018

What's hot (20)

PDF
MyIX Updates by Raja Mohan
PDF
TIME Journey to the SPACE
PDF
MyIX Updates
DOC
Future of broadband in india
PDF
High Speed Fiber Services and Challenges to the Core Network by Seiichi Kawamura
PDF
WINS: Peering and IXPs
PDF
Using LTE to Boost ARPU
PDF
Simplifying IMS - IMS, VoLTE, RCS and LTE
PDF
The Path to a Programmable Network
PDF
PDF
PLNOG 8: Robert Shore - Dynamic Optical Networking
PDF
Scaling Beyond 100G With 400G and 800G
PDF
LTE-Broadcast: Reality Check
PDF
1000x: Higher Efficiency
PDF
M2M-IoT towards 5G
PDF
5G: A 2020 Vision
PDF
Mobile IPv6 course at CACIC 2006
PPTX
Internet Peering and the Role of an IXP
PDF
Multicast QUIC for video content delivery
MyIX Updates by Raja Mohan
TIME Journey to the SPACE
MyIX Updates
Future of broadband in india
High Speed Fiber Services and Challenges to the Core Network by Seiichi Kawamura
WINS: Peering and IXPs
Using LTE to Boost ARPU
Simplifying IMS - IMS, VoLTE, RCS and LTE
The Path to a Programmable Network
PLNOG 8: Robert Shore - Dynamic Optical Networking
Scaling Beyond 100G With 400G and 800G
LTE-Broadcast: Reality Check
1000x: Higher Efficiency
M2M-IoT towards 5G
5G: A 2020 Vision
Mobile IPv6 course at CACIC 2006
Internet Peering and the Role of an IXP
Multicast QUIC for video content delivery
Ad

Similar to Measuring the Internet Economy: How Networks Create Value (20)

PDF
Better Than Best Effort at Bloomberg from ThousandEyes Connect
PDF
Cable Europe factsheet - Internet and traffic management
PDF
From Stream to Screen: Real-Time Data Streaming to Web Frontends with Conflue...
PDF
NetIX - Moving Traffic to the Expressway Jan 2016
PDF
A novel token based approach towards packet loss
PDF
A novel token based approach towards packet loss control
PDF
uMeter QoS | Case Study
PDF
PLNOG16: Public IX is the tip of the Internet Iceberg. The 9:1 PNI rule, Mart...
PDF
PLNOG 17 - Stefan Meinders - Slow is the new Down
PDF
TADSummit EMEA 2019, Challenges Consuming Programmable Telecoms from the Deve...
PDF
Challenges Consuming Programmable Telecoms from the Developer’s Perspective
PDF
PacNOG 31: Internet Exchange Points
PDF
PITA 27th AGM & Business Forum Expo 23: Internet Exchange Points
PDF
BBL AXA Lille - Nearable and the Eddystone Quest
PDF
Peering 101 - ABQNOG1 - May2023
PDF
Community Networks: An Alternative Paradigm for Developing Network Infrastruc...
PPTX
Chapter04
PDF
Arcus Advisors Report_Quality of Service
PDF
Final Interconnection & Termination Rates Programme
PDF
NetIX Moving Traffic to the Expressway 2016
Better Than Best Effort at Bloomberg from ThousandEyes Connect
Cable Europe factsheet - Internet and traffic management
From Stream to Screen: Real-Time Data Streaming to Web Frontends with Conflue...
NetIX - Moving Traffic to the Expressway Jan 2016
A novel token based approach towards packet loss
A novel token based approach towards packet loss control
uMeter QoS | Case Study
PLNOG16: Public IX is the tip of the Internet Iceberg. The 9:1 PNI rule, Mart...
PLNOG 17 - Stefan Meinders - Slow is the new Down
TADSummit EMEA 2019, Challenges Consuming Programmable Telecoms from the Deve...
Challenges Consuming Programmable Telecoms from the Developer’s Perspective
PacNOG 31: Internet Exchange Points
PITA 27th AGM & Business Forum Expo 23: Internet Exchange Points
BBL AXA Lille - Nearable and the Eddystone Quest
Peering 101 - ABQNOG1 - May2023
Community Networks: An Alternative Paradigm for Developing Network Infrastruc...
Chapter04
Arcus Advisors Report_Quality of Service
Final Interconnection & Termination Rates Programme
NetIX Moving Traffic to the Expressway 2016
Ad

More from Bangladesh Network Operators Group (20)

PDF
DNS & DNSSEC operational best practices - Sleep better at night with KINDNS i...
PDF
IPv6 Mostly Experience at APRICOT by Yoshinobu Matsuzaki (IIJ)
PDF
Fast Reroute in SR-MPLS by Md Abdullah Al Naser
PDF
DDoS Mitigation Strategies by Md. Abdul Awal
PDF
DNS & DNSSEC operational best practices - Sleep better at night with KINDNS i...
PDF
Optics101 for non-Optical (IP) folks by Tashi Phuntsho
PPTX
The Internet Service Providers and Connectivity Providers of ICANN
PPTX
Integration of AI and GenAI in Education and beyond
PPTX
Strengthening Cyber Security with Tools and Human Expertise
PDF
Mental Health and Workplace Culture in Tech:A Personal Perspective
PDF
Network Efficiency:The LLM Advantage on network infrastructures
PDF
Utilizing Free and open-source Technology and Achieve Next Generation Enterpr...
PPTX
BDNOG17 Plenary Session, Security Concerns: A perspective in Smart Bangladesh
PPTX
Maximizing Network Efficiency with Large Language Models (LLM)
PPTX
Geolocation and Geofeed Implementation bdNOG18
PDF
Data Centre Design Consideration for Bangladesh
PDF
DNS Troubleshooting - Assumptions and Problem Breakdown
PPTX
Team Cymru Community Services,Overview of all public services
PPTX
Open Source TCP or Netflow Log Server Using Graylog
PPTX
Enhancing seamless access using TIGERfed
DNS & DNSSEC operational best practices - Sleep better at night with KINDNS i...
IPv6 Mostly Experience at APRICOT by Yoshinobu Matsuzaki (IIJ)
Fast Reroute in SR-MPLS by Md Abdullah Al Naser
DDoS Mitigation Strategies by Md. Abdul Awal
DNS & DNSSEC operational best practices - Sleep better at night with KINDNS i...
Optics101 for non-Optical (IP) folks by Tashi Phuntsho
The Internet Service Providers and Connectivity Providers of ICANN
Integration of AI and GenAI in Education and beyond
Strengthening Cyber Security with Tools and Human Expertise
Mental Health and Workplace Culture in Tech:A Personal Perspective
Network Efficiency:The LLM Advantage on network infrastructures
Utilizing Free and open-source Technology and Achieve Next Generation Enterpr...
BDNOG17 Plenary Session, Security Concerns: A perspective in Smart Bangladesh
Maximizing Network Efficiency with Large Language Models (LLM)
Geolocation and Geofeed Implementation bdNOG18
Data Centre Design Consideration for Bangladesh
DNS Troubleshooting - Assumptions and Problem Breakdown
Team Cymru Community Services,Overview of all public services
Open Source TCP or Netflow Log Server Using Graylog
Enhancing seamless access using TIGERfed

Recently uploaded (20)

PPT
415456121-Jiwratrwecdtwfdsfwgdwedvwe dbwsdjsadca-EVN.ppt
PDF
FINAL CALL-6th International Conference on Networks & IOT (NeTIOT 2025)
PDF
Introduction to the IoT system, how the IoT system works
PPTX
Slides PPTX: World Game (s): Eco Economic Epochs.pptx
DOC
Rose毕业证学历认证,利物浦约翰摩尔斯大学毕业证国外本科毕业证
PPTX
artificialintelligenceai1-copy-210604123353.pptx
PPTX
Introduction to cybersecurity and digital nettiquette
PDF
The New Creative Director: How AI Tools for Social Media Content Creation Are...
PPT
FIRE PREVENTION AND CONTROL PLAN- LUS.FM.MQ.OM.UTM.PLN.00014.ppt
PPTX
SAP Ariba Sourcing PPT for learning material
PDF
Smart Home Technology for Health Monitoring (www.kiu.ac.ug)
PDF
Uptota Investor Deck - Where Africa Meets Blockchain
PPTX
E -tech empowerment technologies PowerPoint
PPT
isotopes_sddsadsaadasdasdasdasdsa1213.ppt
PDF
Exploring VPS Hosting Trends for SMBs in 2025
PPTX
t_and_OpenAI_Combined_two_pressentations
PPTX
Internet Safety for Seniors presentation
PPTX
Layers_of_the_Earth_Grade7.pptx class by
PPTX
Funds Management Learning Material for Beg
PDF
Session 1 (Week 1)fghjmgfdsfgthyjkhfdsadfghjkhgfdsa
415456121-Jiwratrwecdtwfdsfwgdwedvwe dbwsdjsadca-EVN.ppt
FINAL CALL-6th International Conference on Networks & IOT (NeTIOT 2025)
Introduction to the IoT system, how the IoT system works
Slides PPTX: World Game (s): Eco Economic Epochs.pptx
Rose毕业证学历认证,利物浦约翰摩尔斯大学毕业证国外本科毕业证
artificialintelligenceai1-copy-210604123353.pptx
Introduction to cybersecurity and digital nettiquette
The New Creative Director: How AI Tools for Social Media Content Creation Are...
FIRE PREVENTION AND CONTROL PLAN- LUS.FM.MQ.OM.UTM.PLN.00014.ppt
SAP Ariba Sourcing PPT for learning material
Smart Home Technology for Health Monitoring (www.kiu.ac.ug)
Uptota Investor Deck - Where Africa Meets Blockchain
E -tech empowerment technologies PowerPoint
isotopes_sddsadsaadasdasdasdasdsa1213.ppt
Exploring VPS Hosting Trends for SMBs in 2025
t_and_OpenAI_Combined_two_pressentations
Internet Safety for Seniors presentation
Layers_of_the_Earth_Grade7.pptx class by
Funds Management Learning Material for Beg
Session 1 (Week 1)fghjmgfdsfgthyjkhfdsadfghjkhgfdsa

Measuring the Internet Economy: How Networks Create Value

  • 1. Measuring the Internet Economy: How Networks Create Value Bill Woodcoc k Executive Directo r Packet Clearing House Version 1. 0 bdNOG1 3 June 10-12, 2021
  • 2. Why Measure the Internet? Tactical reasons: To understand where the value of the product you’re selling comes from As input to calculating Average Per-Bit Delivery Costs (APBDC) So that you can bill your customers in di erent and more useful ways To optimize and plan network interconnection (peering and transit) Strategic reasons: So you’ll know your business when you talk with your regulator or customers To justify infrastructural investment and attention from policy-makers
  • 3. Why Measure the Internet? Although economists argue about methodologies and speci cs, there’s general agreement that the Internet is responsible for a quarter to a third of new productivity in most OECD countries, and a double-digit percentage globally. It’s the responsibility of governmental policy-makers, of regulators, and of people in the involved industries, to understand and support such engines of growth. The Internet allows people without work to fi nd jobs, children to continue their education during a pandemic, businesses to receive orders and ful ll them. The economic activity produces tax revenue which pays for schools, roads, law enforcement, and healthcare. Regardless of whether individual customers appreciate the work that you do, society and posterity depend upon your work and its success.
  • 4. Theodore Vail used the Network E ff ect to justify an AT&T monopoly in 1908. Value increases as a function of the number of users of a compatible product or service. In communications, value increases as a function of the number of other people who can be reached through a network. The Network Effect
  • 5. The Network Effect Users: 2 Value: 1 Value per user: 0.5 Marginal value of next user: 2 Users: 4 Value: 6 Value per user: 1.5 Marginal value of next user: 4 Users: 6 Value: 15 Value per user: 2.5 Marginal value of next user: 6 Users: 5 Value: 10 Value per user: 2 Marginal value of next user: 5 Users: 3 Value: 3 Value per user: 1 Marginal value of next user: 3
  • 7. The Network Effect Network e ff ects apply not only to underlying communications infrastructure like ber cables, but at all layers of the network. Some examples: Layer 2: Ethernet adjacencies Layer 3: Global reachability with the Internet Protocol Layer 5: BGP route-servers like those which allowed the IXPs in Jakarta and Sao Paulo to so quickly outpace those of Europe and the United States Layer 8: Speakers of a common language, or people who share a common understanding of a scienti fi c principle
  • 8. Internet Service Provider Lifecycle Phase 1: Simple Aggregator Single Transit Provider Customers ISP Network IXPs
  • 9. Internet Service Provider Lifecycle Phase 1: Simple Aggregator Single Transit Provider Customers ISP Network IXPs Vertical connections are transit: service travels downward, money travels upward.
  • 10. Redundant Transit Providers Customers ISP Network IXPs Internet Service Provider Lifecycle Phase 2: Redundancy & Efficiency
  • 11. Redundant Transit Providers Customers ISP Network Single IXP IXPs Internet Service Provider Lifecycle Phase 3: Local Peering
  • 12. Redundant Transit Providers Customers ISP Network Single IXP IXPs Horizontal connections are peering: bandwidth is exchanged, but money is not. Internet Service Provider Lifecycle Phase 3: Local Peering
  • 13. Redundant Transit Providers Customers ISP Network Multipl e IXPs IXPs Internet Service Provider Lifecycle Phase 4: Diverse Peering
  • 15. IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1
  • 16. IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1
  • 17. IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1
  • 18. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1
  • 19. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1
  • 20. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1 C C
  • 21. C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 1 1 1 2 C C 1
  • 22. C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 3 2 2 2 1 1 1 3 2 4 C C 1
  • 23. C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP 4 4 3 2 2 2 1 1 1 3 2 IXP 4 C C 1
  • 24. C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP 4 3 2 2 2 1 1 1 3 2 4 IXP 4 C C 1
  • 25. C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP 4 3 2 2 2 1 1 1 2 3 4 IXP 4 C C 1
  • 26. C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP 4 3 2 2 2 1 1 1 3 4 IXP 4 C C 1 2
  • 27. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1
  • 28. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1 Transit Cone Customers: 3 Value: 6
  • 29. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1 Transit Cone Customers: 5 Value: 15
  • 30. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1 Transit Cone Customers: 9 Value: 45
  • 31. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1 Transit Cone Customers: 16 Value: 136
  • 32. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1 Customers: 24 Value: 300 Transit Cone + Peering
  • 34. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1
  • 35. C C C C C C C C C C C C C C C C C C C C C C C C IXP IXP IXP IXP IXP 4 4 4 3 3 2 2 2 2 1 1 1 1 IXP IXP IXP C C C C C C
  • 36. So What Do We Measure? The vast majority of bandwidth is produced in Internet exchange points. Bandwidth produced in IXPs comes from interconnections between participating networks across the switch fabric, or across passive ber crossconnects. Aggregate bandwidth produced on the switch fabric is typically a public number. The amount of bandwidth produced on private pieces of ber is known only to the two parties to each interconnection, but traceroutes will reveal whether the connection is occurring across the switch fabric or a private connection. A tiny remaining fraction of bandwidth is produced within transit cones, and traceroute analysis can identify this, but not quantify it, as well.
  • 37. So What Do We Measure? The ratio of bandwidth produced through each of these three mechanisms is not directly measurable, for many reasons. The ratio of interconnections facilitated by each of the three mechanisms can be determined by analyzing traceroutes, but each of the three types have characteristically di ff erent tra ffi c capacities and average utilization. Di ff erent business practices and regulatory environments strongly a ect the ratios of tra ffi c generated using each mechanism: • Regulatorily-protected monopoly incumbents in many developing economies • Data-collection-driven preference for switching over crossconnects
  • 38. So What Do We Measure? More speci fi cally, at exchange points, we sum the count of outbound bits across all ISP-facing switch ports. Not inbound, because that wouldn’t catch multicast, but would catch bogons and subsequently-mitigated DDoS. Not all ports, because that would double-count inter-switch tra c. If we want to get really fancy, we can deduplicate broadcast tra c, but there really shouldn’t be broadcast tra ffi c on an IX subnet.
  • 39. So What Do We Measure? In an ISP network, measurement is used for both engineering and billing. We typically use fl ow-based measurement, in addition to summing bits in both directions at the peering and upstream transit edges. This allows a “local tra ffi c is free, international tra c is metered” two- category billing model. If you capture customer-facing port bit-counts as well, that enables a “on- net, peer, and transit” three-category model. Old-style customer-port-bit-count-only doesn’t allow di erentiated pricing, nor does it tell you which customers are costing you most.
  • 40. What Questions Can We Answer? If we measure Internet tra ffi c comprehensively, ideally using both bit-counts and sparsely-sampled fl ows, we can answer many questions: • Which of our customers make us pro fi table, and which cost us money? • Is our country or region a net importer or net exporter of bandwidth? • Where is the bandwidth our users are consuming produced? • Where are the consumers of the bandwidth we’re producing? • Which protocols and applications are gaining momentum, and which are falling into disuse?
  • 41. Thanks, and Questions? Bill Woodcoc k Executive Directo r Packet Clearing Hous e woody@pch.ne t https://pch.net