Transcript slides - Caida

Provider and Peer Selection in the
Evolving Internet Ecosystem
Amogh Dhamdhere
Committee:
Dr. Constantine Dovrolis (advisor)
Dr. Mostafa Ammar
Dr. Nick Feamster
Dr. Ellen Zegura
Dr. Walter Willinger (AT&T Labs-Research)
The Internet Ecosystem


27,000 autonomous networks independently operated
and managed
The “Internet Ecosystem”



Network interactions



Localized, in the form of interdomain links
Competitive (customer-provider), symbiotic (peering)
Distributed optimizations by each network


Different types of networks
Interact with each other and with “environment”
Select providers and peers to optimize utility function
The Internet ecosystem evolves
4/13/2015
2
High Level Questions


How does the Internet ecosystem evolve?
What is the Internet heading towards?





Topology
Economics
Performance
How do the strategies used by networks affect their
utility (profits/costs/performance)?
How do these individual strategies affect the global
Internet?
4/13/2015
3
Previous Work

Static graph properties




Homogeneity


Match graph properties
e.g. degree distribution
Nodes and links all the
same
Game theoretic,
computational

4/13/2015
Restrictive assumptions
Dynamics of the evolving
graph

“Descriptive” modeling


No focus on how the graph
evolves

“Bottom-up”


Birth/death
Rewiring
Model the actions of
individual networks
Heterogeneity


Networks with different
incentives
Semantics of interdomain
links
4
Our Approach
“Measure – Model – Predict”
Measure the evolution
Model strategies
Predict the effects
of the Internet
and incentives of
of provider and
Ecosystem
different network
peer selection
Topological, but focus
types
strategies
on business types and
rewiring
4/13/2015
5
Outline

Measuring the Evolution of the Internet Ecosystem
[IMC ’08]


The Core of the Internet: Provider and Peer
Selection for Transit Providers
[to be submitted]
The Edge of the Internet: ISP and Egress Path
Selection for Stub Networks
[Infocom ‘06]

ISP Profitability and Network Neutrality
[Netecon ’08]
4/13/2015
6
Motivation





How did the Internet ecosystem evolve during the
last decade?
Is growth more important than rewiring?
Is the population of transit providers increasing or
decreasing?
 Diversification or consolidation of transit market?
Given that the Internet grows in size, does the
average path length also increase?
Where is the Internet heading?
4/13/2015
7
Approach

Focus on Autonomous Systems (ASes)


Start from BGP routes from RouteViews and RIPE
monitors during 1997-2007


Focus on primary links
Classify ASes based on their business function


As opposed to networks without AS numbers
Enterprise ASes, small transit providers, large transit
providers, access providers, content providers
Classify inter-AS relations as “transit” and “peering”
 Transit link – Customer pays provider
Visibility Issue

4/13/2015
Peering link – No money exchanged
8
Internet growth



Number of CP links and ASes showed initial exponential growth
until mid-2001
Followed by linear growth until today
Change in trajectory followed stock market crash in North
America in mid-2001
4/13/2015
9
Path lengths stay constant

Number of ASes has grown from 5000 in 1998 to 27000 in 2007

Average path length remains almost constant at 4 hops
4/13/2015
10
Rewiring more important than
growth


Most new links due to internal rewiring and not birth (75%)
Most dead links are due to internal rewiring and not death
(almost 90%)
4/13/2015
11
Classification of ASes based on
business function

Four AS types:






Enterprise customers (EC)
Small Transit Providers
(STP)
Large Transit Providers
(LTP)
Content, Access and Hosting
Providers (CAHP)
Based on customer and peer
degrees
Classification based on
decision-trees

80-85% accurate
4/13/2015
12
Evolution of AS types



LTPs: constant population (top-30 ASes in terms of customers)
Slow growth of STPs (30% increase since 2001)
EC and CAHP populations produce most growth

4/13/2015
Since 2001: EC growth factor 2.5, CAHP growth factor 1.5
13
Multihoming by AS types



CAHPs have increased their multihoming degree significantly
 On the average, 8 providers for CAHPs today
Multihoming degree of ECs almost constant (average < 2)
Densification of the Internet occurs at the core
4/13/2015
14
Conjectures on the Evolution of
peering

Peering by CAHPs has increased significantly

4/13/2015
CAHPs try to get close to sources/destinations of content
15
Conclusions
Where is the Internet heading?






Initial exponential growth up to mid-2001, followed
by linear growth phase
Average path length practically constant
Rewiring more important than growth
Need to classify ASes according to business type
ECs contribute most of the overall growth
Increasing multihoming degree for STPs, LTPs and
CAHPs


Densification at the core
CAHPs are most active in terms of rewiring, while ECs
are least active
4/13/2015
16
Outline




Measuring the Evolution of the Internet Ecosystem
The Core of the Internet: Provider and Peer
Selection for Transit Providers
The Edge of the Internet: ISP and Egress Path
Selection for Stub Networks
ISP Profitability and Network Neutrality
4/13/2015
17
Modeling the Internet Ecosystem

From measurements: Significant rewiring activity


Networks rewire connectivity to optimize a certain
objective function




Especially by transit providers
Distributed
Localized spatially and temporally
Rewiring by changing the set of providers and peers
What are the global, long-term effects of these
distributed optimizations?



4/13/2015
Topology and traffic flow
Economics
Performance (path lengths)
18
The Feedback Loop
Interdomain
TM
Interdomain
topology
Routing
Traffic
flow
Cost/price
parameters
Per-AS
profit
Provider
selection
When
Peer
selection
does it converge?
When no network has the incentive to
change its connectivity – “steady-state”
4/13/2015
19
Impact of provider/peer selection
strategies
P1
C
P1P2
open
to
and
P3
No
peering
peering
peer with
CPs
C
P2
C
4/13/2015
S
P3
C
S
S
S
20
Our Approach


What is the outcome when networks use certain
provider and peer selection strategies?
Model the feedback loop in the Internet ecosystem





Computationally find a “steady-state”


Real-world economics of transit, peering, operational costs
Realistic routing policies
Geographical constraints
Provider and peer selection strategies
No network has further incentive to change connectivity
Measure properties of the steady-state

4/13/2015
Topology, traffic flow, economics
21
Network Types

Enterprise Customers (EC)



Small Transit Providers (STP)



Provide Internet transit
Mostly regional in presence (e.g. France Telecom)
Large Transit Providers (LTP)


Stub networks at the edge (e.g. Georgia Tech)
Either sources or sinks
Transit providers with global presence (e.g. AT&T)
Content Providers (CP)
Provider
and of
peer
selection
for STPs and LTPs
 Major sources
content
(e.g. Google)
4/13/2015
22
What would happen if..?






The traffic matrix consists of mostly P2P traffic?
P2P traffic benefits STPs, can make LTPs
unprofitable
LTPs peer with content providers?
LTPs could harm STP profitability, at the expense of
longer end-to-end paths
Edge networks choose providers using path lengths?
LTPs would be profitable and end-to-end paths
shorter
4/13/2015
23
Provider and Peer Selection

Provider selection strategies




Peer selection strategies




Minimize monetary cost (PR)
Minimize AS path lengths weighted by traffic (PF)
Avoid selecting competitors as providers (SEL)
Peer only if necessary to maintain reachability (NC)
Peer if traffic ratios are balanced (TR)
Peer by cost-benefit analysis (CB)
Peer and provider selection are related
4/13/2015
24
Provider and Peer Selection are
Related
A
X
?

C
Restrictive
peering
B
A
B
B
A
C
4/13/2015
C
U

U

Peering by
necessity
Level3-Cogent
peering dispute
25
Economics, Routing and Traffic
Matrix

Realistic transit, peering and operational costs



BGP-like routing policies


Transit prices based on data from Norton
Economies of scale
No-valley, prefer customer, prefer peer routing policy
Traffic matrix



4/13/2015
Heavy-tailed content popularity and consumption by sinks
Predominantly client-server: Traffic from CPs to ECs
Predominantly peer-to-peer: Traffic between ECs
26
Algorithm for network actions


Networks perform their actions sequentially
Can observe the actions of previous networks


Network actions in each move





And the effects of those actions
Pick set of preferred providers
Attempt to convert provider links to peering links “due to
necessity”
Evaluate each existing peering link
Evaluate new peering links
Networks make at most one change to their set of
peers in a single move
4/13/2015
27
Solving the Model



Determine the outcome as each network selects
providers and peers according to its strategy
Too complex to solve analytically: Solve
computationally
Typical computation





4/13/2015
Proceeds iteratively, networks act in a predefined sequence
Pick next node n to “play” its possible moves
Compute routing, traffic flow, AS fitness
Repeat until no player has incentive to move
“steady-state” or equilibrium
28
Properties of the steady-state

Is steady-state always reached?


Is steady-state unique?



Yes, in most cases
No, can depend on playing sequence
Different steady-states have qualitatively similar properties
Multiple runs with different playing sequence


4/13/2015
Average over different runs
Confidence intervals are narrow
29
Canonical Model


Parameterization of the model that resembles real
world
Traffic matrix is predominantly client-server (80%)






Impact of streaming video, centralized file sharing services
20% of ECs are content sources, 80% sinks
Heavy tailed popularity of traffic sources
Edge networks choose providers based on price
5 geographical regions
STPs cheaper than LTPs
4/13/2015
30
Model Validation


Reproduces almost constant average path length
Activity frequency: How often do networks change
their connectivity?

4/13/2015
ECs less active than providers – Qualitatively similar to
measurement results
31
Results – Canonical Model
LTP

Traffic
can bypass LTPs –
LTPs unprofitable
S1
S2
EC
CP
4/13/2015
CP
EC
CP
Hierarchy of STPs
STPs
should not peer
with CPs
Resist the temptation!
32
Results – Canonical Model
LTP
CP
CP
S1
CP
What-if:
CPs
LTPs peer with
Generate
revenue from
downstream traffic
S2
Can
EC
harm STP fitness
EC
Long
4/13/2015
paths
33
Deviation 1: P2P Traffic matrix
LTP
CP
CP
S1
S2
EC
4/13/2015
CP
ECEC
EC
EC
traffic helps STPs
Smaller traffic volume
from CPs to Ecs
More
S3
S2
P2P
EC
EC-EC traffic =>
balanced traffic ratios
More opportunities for
STPs to peer
Peering by “traffic
ratios” makes sense
34
Conclusions


A model that captures the feedback loop between
topology, traffic and fitness in the Internet
Considers effects of




Economics
Geography
Heterogeneity in network types
Predict the effects of provider and peer selection
strategies

4/13/2015
Topology, traffic flow, economics, and performance
35
Outline




Measuring the Evolution of the Internet Ecosystem
The Core of the Internet: Peer and Provider
Selection for Transit Providers
The Edge of the Internet: ISP and Egress Path
Selection for Stub Networks
ISP Profitability and Network Neutrality
4/13/2015
36
The Edge of the Internet

Sources and sinks of content



Content Providers (CP): sources
Enterprise Customers (EC): sinks
From measurements:

ECs connect increasingly to STPs


CPs connect increasingly to LTPs


Performance ?
Increasing multihoming (about 60% of stubs)


Cost conscious ?
Redundancy, load balancing, cost effectiveness
How should stub networks choose their providers?
4/13/2015
37
Major Questions

How to select the set of
upstream ISPs ?




Low monetary cost
Short AS paths to major
destinations
Path diversity to major traffic
destinations – robustness to
network failures
How to allocate egress traffic
to the set of selected ISPs ?


4/13/2015
Objective: Avoid congestion on the
upstream paths
Also maintain low cost
38
ISP Selection

Select k ISPs out of N

Let C be a subset of k ISPs out of N


Total cost of a selection of ISPs C: Weighted sum of
monetary, path length and path diversity costs
Select combination C with minimum total cost

4/13/2015
Feasible to enumerate all combinations
39
Monetary and Path Length Cost


For set of ISPs C, what is the monetary and path
length cost of routing egress flows?
Find the minimum cost mapping G* of flows to ISPs
(Bin Packing)




Use First Fit Decreasing (FFD) heuristic


Flows = items
ISPs = bins
NP hard !
Mapping G* very close to optimal
Monetary and path length costs of C are calculated
using the mapping G*
4/13/2015
40
Path Diversity


selection C gives K paths to
each destination d
K-shared link to d: link shared
by all K paths to d


Minimize the number of Kshared links


If a K-shared link fails,
destination d is unreachable
Path diversity metric: The number
of k-shared links to destination d
averaged over all destinations
Gives the best resiliency to
single-link failures
4/13/2015
41
Summary

Algorithms for ISP selection



ISP selection for monetary and performance
constraints



Choosing best set of upstream ISPs
Objectives are minimum monetary cost, short AS paths and
high path diversity
Formulated as a bin-packing problem
Heuristic gives solution very close to optimal
ISP selection for path diversity

4/13/2015
Returns set of ISPs with best path diversity to the set of
major destinations
42
Outline




Measuring the Evolution of the Internet Ecosystem
The Edge of the Internet: ISP and Egress Path
Selection for Stub Networks
The Core of the Internet: Peer and Provider
Selection for Transit Providers
ISP Profitability and Network Neutrality
4/13/2015
43
The debate


Recent evolution trend: Large amounts of video and
peer-to-peer traffic
Content providers (CP) generate the content



Access Providers (AP) deliver content to users




Provide content and services “over the top” of the basic
connectivity provided by ISPs
Profitable (think Google)
Recent trend: Not profitable
Commoditization of basic Internet access
Want a share of the pie
Tension between AP and CPs: “Network neutrality”
4/13/2015
44
A Technical View

Previous work





Mostly non-technical
Highly emotional debates in the press
Legislation/policy aspects: Do we need network neutrality
legislation?
But what about the underlying problem: Nonprofitability of Access Providers?
Our approach: A quantitative look at AP profitability


4/13/2015
Investigate reasons for non-profitability
Evaluate strategies for remaining profitable
45
Modeling AP Profitability

Three AS types: AP, CP
and transit provider (TP)


AS links



4/13/2015
Focus on the AP
customer-provider
(customer pays provider)
peering (no payments)
AP and CPs can transfer
traffic either through
customer-provider or
peering links
46
AP Profitability

Reasons why APs can be unprofitable



AP strategies: Pricing




AP users
The impact of video traffic
Heavy hitter charging
Heavy hitter blocking
Non-neutral charging
AP strategies: Connection


4/13/2015
Caching CP content
Peering selectively with CPs
47
Major Findings



Variability in AP users can cause large variability in
costs
Video traffic: Increases costs for AP
AP strategies based on differential/non-neutral
pricing may not succeed


Have to account for user departure due to competition
AP strategies based on connection are promising


4/13/2015
Caching content from CPs
Peering selectively with large CPs
48
Contributions of this Thesis


A measurement study of the evolution of the
Internet ecosystem
Modeling the evolution of the Internet ecosystem


Optimizations at the edge of the Internet


“what-if” questions about possible evolution paths
Algorithms for provider selection and egress routing
A technical view of the network neutrality debate

4/13/2015
Strategies for ISP profitability
49
Future Directions

Measurements: Investigate the evolution of the
connectivity for monitor ASes



We can observe all links for such ASes
Focus on transitions between peering and customer-provider
links
Measurements: What does the interdomain traffic
matrix really look like?


4/13/2015
Can we use measurements from a large Tier-1 provider?
Can we augment that data with information about the
interdomain topology?
50
Future Directions

What is the best strategy for different types of
providers?



Strategies for classes of providers
Strategies for individual providers
Do the distributed optimizations by networks solve a
centralized problem?

4/13/2015
E.g., minimizing path lengths
51
Other things I’ve been up to

Router buffer sizing



[CCR ‘06]
“NetDiagnoser: Troubleshooting network disruptions using
[CoNext ‘07]
end-to-end probes and routing data”
Network monitoring


[Infocom ‘05]
Network troubleshooting


“Buffer Sizing For Congested Internet Links”
“Open Issues in Router Buffer Sizing”
“Route monitoring from passive data plane measurements”
Measurement


4/13/2015
“Poisson vs. Periodic Path Probing”
“Bootstrapping in Gnutella”
[In progress]
[IMC ‘05]
[PAM ‘04]
52
Thank You !
4/13/2015
53
Issue-1: remove backup/transient links

Each snapshot of the Internet topology captures 3
months


40 snapshots – 10 years
Perform “majority filtering” to remove backup and
transient links from topology



For each snapshot, collect several “topology samples”
interspersed over a period of 3 weeks
Consider an AS-path only if it appears in the majority of the
topology samples
Otherwise, the AS-path includes links that were active for
less than 11 days (probably backup or transient links)
Samples
Snapshot
54
4/13/2015
4/13/2015
Issue-2: variable set of BGP monitors

Some observed link births may be links revealed due to increased
monitor set


Similarly for observed link deaths
We calculated error bounds for link births and deaths


Relative error < 10% for CP links
See paper for details
4/13/2015
55
Issue-3: visibility of ASes, Customer-Provider (CP) and
Peering (PP) links


Number of ASes and CP links is robust to number of monitors
But we cannot reliably estimate the number of PP links
4/13/2015
56
Global Internet trends
4/13/2015
57
Transit (CP) vs Peering (PP) relations

The fraction of peering links has been increasing steadily


But remember: this is just a lower bound
At least 20% of inter-AS links are of PP type today
4/13/2015
58
The Internet gets larger but not
longer


Average path length remains almost constant at 4 hops
Average multihoming degree of providers increases faster than
that of stubs

4/13/2015
Densification at core much more important than at edges
59
4/13/2015
60
Regional distribution of AS types



Europe is catching up with North America w.r.t the population of
ECs and LTPs
CAHPs have always been more in Europe
More STPS in Europe since 2002
4/13/2015
61
Evolution of Internet transit:
the customer’s perspective
4/13/2015
62
Customer activity by region


Initially most active customers were in North America
After 2004-05, customers in Europe have been more active


4/13/2015
Due to increased availability of providers?
More competitive market?
63
How common is multihoming among AS
species?

CAHPs have increased their multihoming degree significantly



On the average, 8 providers for CAHPs today
Multihoming degree of ECs has been almost constant (average < 2)
Densification of the Internet occurs at the core
4/13/2015
64
Who prefers large vs small transit
providers?

After 2004, ECs prefer STPs than LTPs


Mainly driven by lower prices or regional constraints?
CAHPs connect to LTPs and STPs with same probability
4/13/2015
65
Customer activity by region


Initially most active customers were in North America
After 2004-05, customers in Europe have been more active


4/13/2015
Due to increased availability of providers?
More competitive market?
66
Evolution of Internet transit:
the provider’s perspective
4/13/2015
67
Attractiveness (repulsiveness) of transit providers

Attractiveness of provider X: fraction of new CP links that connect to X



Repulsiveness, defined similarly
Both metrics some positive correlation with customer degree
Preferential attachment and preferential detachment of rewired links
4/13/2015
68
Evolution of attractors and
repellers


A few providers (50-60) account for 50% of total
attractiveness (attractors)
The total number of attractors and repellers increases



The Internet is NOT heading towards oligopoly of few large players
LTPs dominate set of attractors and repellers
CAHPs are increasingly present however
4/13/2015
69
Correlation of attractiveness and
repulsiveness



Timeseries of attractiveness and repulsiveness for each provider
Calculate cross-correlation at different lags
Most significant correlation values at lags 1,2 and 3

4/13/2015
Attractiveness precedes repulsiveness by 3-9 months
70
Evolution of Internet peering
(conjectures)
4/13/2015
71
Evolution of Internet Peering


ECs and STPs have low peering frequency
Aggressive peering by CAHPs after 2003

4/13/2015
Open peering policies to reduce transit costs
72
Which AS pairs like to peer?

Peering by CAHPs has increased significantly


CAHPs try to get close to sources/destinations of content
Peering by LTPs has remained almost constant (or declined)

4/13/2015
“Restrictive” peering by LTPs
73
Conclusions
Where is the Internet heading towards?






Initial exponential growth up to mid-2001, followed
by linear growth phase
Average path length practically constant
Rewiring more important than growth
Need to classify ASes according to business type
ECs contribute most of the overall growth
Increasing multihoming degree for STPs, LTPs and
CAHPs


Densification at core
CAHPs are most active in terms of rewiring, while ECs
are least active
74
4/13/2015
4/13/2015
Conclusions
Where does the Internet head toward?




Positive correlations between attractiveness &
repulsiveness of provider and its customer degree
Strong attractiveness precedes strong repulsiveness
by period of 3-9 months
Number of attractors and repellers between shows
increasing trend
The Internet market will soon be larger in Europe
than in North America


In terms of number of transit providers and CAHPs
Providers from Europe increasingly feature in the set
of attractors and repellers
4/13/2015
75
Multihoming

Multihoming: Connection of a
stub network to multiple ISPs




Redundancy

primary/backup relationships

Distribute outgoing traffic among
ISPs
Load Balancing
Cost Effectiveness


x% of stub networks are
multihomed
Lower cost ISP for bulk traffic,
higher cost ISP for performancesensitive traffic
Performance

4/13/2015
Intelligent Route Control
76
ISP selection


Should consider both monetary cost and
performance
Minimum monetary cost


Minimum AS path lengths



Estimate the cost that “would be” incurred if a set of
ISPs was selected
Longer paths: delays, interdomain routing failures
Measure AS path length offline using Looking Glass
Servers
Maximum Path diversity

4/13/2015
AS-level paths to destinations should be as “different”
as possible
77
Problem Definition


Two phases
Phase I – ISP Selection:





Select K upstream ISPs
K depends on monetary and performance constraints
“Static” operation
Change only when major changes in the traffic destinations or
ISP pricing
Phase II – Egress Path Selection



4/13/2015
Allocate egress traffic to selected ISPs
Avoid long term congestion and minimize cost
“Semi-static” operation, performed every few hours or days
78
Evaluation – Path Diversity

AS-level paths and traffic
rates are input to simulator




9 ISPs, 250 destinations
Given K, find the selection
C* with the minimum path
diversity cost
For each selection C, find
u(C) = total traffic lost due
to the failure of each link in
topology
Calculate Δu(C) = u(C) – u(C*)
for each selection C
4/13/2015

Single link failures:
C* is the optimal
selection
79
Egress Path Selection



After Phase-I, S has K upstream ISPs
Problem: How to map outgoing traffic to the ISPs
M flows: KM mappings of flows to ISPs




Some mappings may cause congestion to flows !
Flows can be congested at access links or further upstream
Objective: Find the loss-free mapping with the
minimum cost
Challenges:


Upstream topology and capacities are unknown
Iterative routing approaches required

4/13/2015
Propose an iterative routing based on simulated annealing
80
Evaluation – Path Diversity

AS-level paths and traffic
rates are input to simulator




9 ISPs, 250 destinations
Given K, find the selection
C* with the minimum path
diversity cost
For each selection C, find
u(C) = total traffic lost due
to the failure of each link in
topology
Calculate Δu(C) = u(C) – u(C*)
for each selection C
4/13/2015


Single
failures:
2,3 linklink
failures:
C*
C* is
is the
closeoptimal
to the
selection
optimal selection
81
Provider and Peer Selection

Detailed model for provider and peer selection


Provider selection objectives



Monetary cost
AS path lengths
Peer selection



Complex real-world decisions
Minimize transit costs
Maintain reachability
Constraints


4/13/2015
Only local knowledge
Geographical constraints
82
Peering Federation
A



C
Traditional peering links: Not transitive
Peering federation of A, B, C: Allows mutual transit


B
Longer chain of “free” traffic
Incentives to join peering federation?
What happens to tier-1 providers if smaller providers form
federations?
4/13/2015
83
Why can the AP be unprofitable?



Variability of users =>
high variability in the
costs incurred by AP
Variability increases
with the access speed
Increase in video
traffic: higher transit
payment by AP
4/13/2015
84
Baseline model



AP and CP connect to the TP as customers
N users of AP, charged a flat rate R ($/month)
Transit pricing: 95th percentile of traffic volume,
concave transit pricing functions



95th / mean = 2:1 for normal traffic, 4:1 for video1
More video means higher transit payment by AP
AP users: Heavy tailed distribution of content
downloaded per month

High variability in AP costs
1Norton’06:
4/13/2015
Internet Video: The Next Wave of Massive Disruption to the U.S. Peering Ecosystem
85
AP Strategies

Charging strategies




AP charges “heavy hitters”
according to volume
downloaded
AP caps heavy hitters
AP charges CP (non-network
neutral)
Charging strategies are
disruptive

AP cannot control customer
departure probability
4/13/2015
86
AP Strategies

Charging heavy hitters




download amount D,
threshold T, flat rate R
c(D) = D*R/T
AP’s profit is sensitive to
customer departure prob
Capping heavy hitters
and non-neutral charging
would not work for the
same reason
4/13/2015
87
AP Strategies

Connection Strategies




Non-disruptive
Caching can reduce transit costs of AP


AP caches content from CPs
AP peers with CPs
But depends on the amount of content cacheable
Selective peering with CPs can improve profitability


Peering cost depends on CP
Cost/benefit analysis for each CP

4/13/2015
CP with large network: low cost of peering
88
AP Strategies

Connection Strategies




Non-disruptive
Cost-benefit analysis for
peering




AP caches content from CPs
AP peers with CPs
Peering cost depends on CP
(easy/medium/hard)
r = saving/cost (both
estimated)
Peer if r > R
AP controls the factor R
4/13/2015
89
AP Strategies

Charging heavy hitters




download amount D,
threshold T, flat rate R
c(D) = D*R/T
AP’s profit is sensitive to
customer departure prob
Non-neutral charging



4/13/2015
Customer departure prob
“How discriminatory is my
AP?”
AP’s profit is sensitive to
customer departure prob
90
Why Study Internet Evolution?

“Bottom-up” models (more later)


Performance of protocols over time


Understand how local actions lead to emerging properties
“How would BGP perform 10 years from now?”
Clean slate vs. evolutionary design


4/13/2015
After initial design, both must evolve !
Understanding evolution of the current Internet can help
design
91