Download - GENESIS: An agent-based model of interdomain network formation, traffic flow and economics
![Page 1: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/1.jpg)
1
GENESIS: An agent-based model of interdomain network formation,
traffic flow and economics
Aemen Lodhi (Georgia Tech)Amogh Dhamdhere (CAIDA)
Constantine Dovrolis (Georgia Tech)
31st Annual IEEE International Conference on Computer Communications (IEEE INFOCOM 2012)
![Page 2: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/2.jpg)
2
Outline• GENESIS: Introduction & Motivation• The model: Key features• Results– Validation– Analysis of results
• Case study• How to use GENESIS in your research
![Page 3: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/3.jpg)
3
INTRODUCTION
![Page 4: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/4.jpg)
4
Motivations for an interdomain network formation model
• Insight into dynamics of interdomain network
• Study pricing schemes• Study increasing asymmetry in
interdomain traffic matrix• Evaluate peering strategies• Impact of actions on economic fitness• Internet “ecosystem” in the future?
![Page 5: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/5.jpg)
5
What is GENESIS• Agent based interdomain network
formation model• Autonomous Systems (AS) as
independent agents acting in a distributed asynchronous manner
Enterprise
customer
Transit Provide
r
Transit Provide
r
Internet
Enterprise
customerContent Provider
Content Provider
![Page 6: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/6.jpg)
6
What is GENESIS• Actions by ASes– Transit provider selection– Peering strategy selection– Peering and Depeering decisions
• Outcome of these actions– Formation of an interdomain network
starting from a random initial state–Mostly ending in equilibrium
![Page 7: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/7.jpg)
7
What GENESIS is not• Not a topology generation model• Not a crystal ball to accurately
predict the economic fitness or hierarchical status of a single specific AS in future
• Use GENESIS for – computing statistical properties of
network topology + economic fitness of different categories of ASes
![Page 8: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/8.jpg)
8
THE MODEL
![Page 9: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/9.jpg)
9
Model features• Geographic co-location constraints in
provider/peer selection• Traffic matrix• Public & Private peering• Set of peering strategies• Transit provider selection mechanism• Economic attributes: Peering costs,
Transit costs, Transit revenue
![Page 10: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/10.jpg)
Model features
Fitness = Transit Revenue – Transit Cost – Peering cost
• Objective: Maximize economic fitness• Optimize connectivity through peer
and transit provider selection
![Page 11: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/11.jpg)
11
Geographic presence & constraints
Regions corresponding to unique
IXPs
Geographic overlap
![Page 12: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/12.jpg)
12
Traffic Matrix• Traffic for ‘N’’ size network represented
through an N * N matrix• Illustration of traffic matrix for a 4 AS
network
00
00
30
20
10
030201
ttt
ttt
Traffic sent by AS 0 to other ASes in
the network
Traffic received by AS 0 from
other ASes in the network
Intra-domain traffic not
captured in the model
![Page 13: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/13.jpg)
13
Traffic components
Autonomous system
Inbound traffic Traffic
consumed in the AS
Traffic transiting
through the AS
Traffic generated
within the AS
Outbound traffic
• Transit traffic = Inbound traffic – Consumed trafficsame as
• Transit traffic = Outbound traffic – Generated traffic
![Page 14: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/14.jpg)
14
Peering strategies• Restrictive: Peer only to avoid network
partitioning• Selective: Peer with ASes of similar
size
• Open: Every co-located AS except customers
![Page 15: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/15.jpg)
15
Peering strategy selection
• Default model– Tier 1 Transit providers: Restrictive– All other transit providers: Selective– Stubs: Open
![Page 16: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/16.jpg)
16
Execution of a sample path
1 2 N
Iteration
1. Depeering2. Peering3. Transit provider
selection4. Peering strategy
update
1. Depeering2. Peering3. Transit provider
selection4. Peering strategy
update
1. Depeering2. Peering3. Transit provider
selection4. Peering strategy
update
1 2 N
Iteration
Time
• No exogenous changes• Finite states
![Page 17: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/17.jpg)
17
RESULTS
![Page 18: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/18.jpg)
18
Stability of the model
1 2 N
Iteration
1 2 N
Iteration
Time
• Equilibrium: No topology, peering strategy changes in two consecutive iterations
• 90% simulations reach equilibrium• Short time scales• Average time to equilibrium: 6
iterations
![Page 19: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/19.jpg)
19
Oscillations: An artifact?
1 2 N
Iteration
1 2 N
Iteration
Time
• 10% simulations oscillate• Always involve Tier-1 ASes• Resemble real Tier-1 peering disputes• GENESIS captures that endogenous
dynamics cannot always produce stable network
• Exogenous intervention required
![Page 20: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/20.jpg)
20
Validation• Comprehensive validation not possible• Should be viewed as proof of concept• 10% ASes end up being transit providers• Average path length 3.7 (500 nodes) vs.
Average Internet measured path length 4• Path length does not increase significantly
as GENESIS scales from 500 to 1000 nodes
![Page 21: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/21.jpg)
21
Validation• Highly skewed degree distribution• Not exactly a power law owing to limited
number of nodes• Few links carry several orders of magnitude
more traffic
![Page 22: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/22.jpg)
22
Variability across equilibria
• Sources of variation in a single population: Initial topology, Playing order
• Same population but different initial topology: 85% distinct equilibria
• Same population & initial topology but different playing order: 90% distinct equilibria
• Distinct equilibria quite similar in terms of topology
• Coefficient of variation of fitness close to zero for 90% ASes
![Page 23: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/23.jpg)
23
Variability across equilibria
• Most predictable ASes– Stubs: Enterprise customers, Small ISPs– Very large transit providers
• Most unpredictable ASes–Midsize (regional) transit providers
![Page 24: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/24.jpg)
24
Case study: Peering Openness• How does peering openness affect the
properties of the network?• Optimal fitness in range of peering ratios
observed in the real world (1.5 to 5)
![Page 25: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/25.jpg)
25
Case study: Peering Openness• Widespread peering: Saving on costs
not the only outcome• Results in loss of transit revenue
![Page 26: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/26.jpg)
26
Summary of GENESIS findings
• Individual AS status hard to predict• Regional transit providers most
sensitive to network level changes• Overall network characteristics more
predictable• Internet a stable network (mostly) in
the absence of exogenous factors• Increased peering may result in loss of
transit revenue
![Page 27: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/27.jpg)
How can I use GENESIS in my research?
• Flexible & Modular
27
Peering strategi
es
Resulting network
Traffic matrix
Pricing schemes
Presence at IXPs
Presence at IXPs
Peering strategi
es
![Page 28: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/28.jpg)
28
How can I use GENESIS in my research?
• C++ single thread implementation• Fast: average simulation time for 500 nodes: 1.25
hours• Scales up to 1000 nodes• Used in “Analysis of peering strategy adoption by
transit providers in the Internet” NetEcon 2012• Available at:
www.cc.gatech.edu/~dovrolis/Papers/genesis.zip
![Page 29: GENESIS: An agent-based model of interdomain network formation, traffic flow and economics](https://reader036.vdocuments.us/reader036/viewer/2022070422/56816556550346895dd7d373/html5/thumbnails/29.jpg)
29
THANK YOU