DIMES

1
DIMES

• Distributed Internet MEasurement and Simulation
• S. Solomon (HUJI) E. Shir (Cogniview)

2
The Problem

• Centrally measuring the Internets topology is an
  impossible task, both in the AS and in the router
  level.
• Hidden BGP connections.
• 2. Simulating centrally the Internet is not an
  alternative
• The characteristics of the Internet as a
  spontaneously developed WAN are presently not
  fully understood
• The present machines are not powerful enough.
• Future worse
• Number of nodes O (Exp (time))
• A simulation is as accurate as its underlying
  assumptions (Which in the case of the Internet
  have presently scant experimental basis see 1.)

3
The SolutionThe Net which measures itself The
Net which simulates itselfThe Net which
foretells its own future

• Instead of using a single machine
• to study the Internet
• we will use
• (potentially)
• all the machines on the Internet

4
The Goal

• Creating a distributed platform that will enable
• Global scale measurement of Internet graph
  structure, packet traffic statistics, demography
• Simulation of Internet behavior under different
  conditions
• Simulation of the Internet future
• Active networks
• Novel routing algorithms
• Distributed resource allocation grid computing
• Mobile(/Ad-hoc) networks

5
The method

• Recruit millions of nodes spread over the entire
  Net that
• Perform local measurements of their Internet
  neighborhood
• Participate as message routing/processing
  elements in collective Internet experiments
• A generation beyond SETI_at_Home

6
Deliverables (examples)

• Connectivity statistics (links power law)
  including hidden links 6 months
• Topology (K-Core, small worldness) including
  hidden links 12 months
• Corresponding I/O traffic statistics (power law?)
  18 months
• Usage mode statistics (e.g. HTTP vs. P2P)
• Traffic flow mapping
• Optimization of network design 30 months
• Optimization of network usage 36 months

7
The Architecture
Nodes on the Internet
Mirror Server
Experimental Results
Virtual Router Data Aggregator/ Transmitter Manage
ment/Authentication Engine Visualization
Engine Simulation Engine Broadcast
Engine Super-Node Engine
Experiments Control
Experiments Server
Experiments Conductor Receiver/Broadcaster Visuali
zation Engine Analysis/Data Mining Engine