LiTGen, a lightweight traffic generator: application to mail and P2P wireless traffic

1
LiTGen, a lightweight traffic generator
application to mail and P2P wireless traffic

• Chloé Rolland, Julien Ridoux and Bruno Baynat
• Laboratoire LIP6 CNRS
• Université Pierre et Marie Curie Paris 6
• ARC Special Research Center for Ultra-Broadband
  Communications (CUBIN),
• The University of Melbourne

2
Generating IP traffic with accurate timescales
properties

• General framework multiple applications
• LiTGen, a lightweight traffic generator
• Semantically meaningful structure
• Does not rely on a network and/or TCP emulator
• Fast computation
• Measurement based validation
• Application to mail and P2P wireless traffic

3
LiTGens underlying model

• Focus on the download path
• Do not consider up/down interactions
• Focus on TCP traffic
• Approach
• Application oriented User oriented
• Semantically meaningful hierarchical model

4
LiTGens underlying model
IS
PACKETS
5
Basic vs. Extended LiTGen

• Basic LiTGen
• Renewal processes
• Successive random variables (R.V.) i.i.d.
• No dependency between different R.V.
• Extended LiTGen
• Renewal processes
• Dependency introduced, the average packets
  inter-arrival depends on the objects size IApkt
  f(Nobj)

6
Calibration by inspection of the wireless trace

• Wireless trace US ISP wireless network

7
Validation methodology

• Wavelet analysis of the packets arrival times
  series (LDE)

Energy spectrum comparison
?
Captured trace
Synthetic trace
8
Comparison of different kinds of traffics spectra
(1/2)
Web Mail P2P traffic
9
Comparison of different kinds of traffics spectra
(2/2)
Mail traffic
P2P traffic
10
Further validation semi-experiments (SE)

• Does LiTGen reproduces the traffic internal
  structure?
• Semi-experiments
• Manipulation of internal parameters
• Impact of the manipulation importance of the
  parameters modified ?

11
Example of SE P-Uni

• Uniformly distributes packets arrival times
  within each object
• Examine impact of in-objects packets burstiness

P-Uni
Captured trace
P-Uni
Synthetic trace
12
SE results mail traffic
Captured trace
Synthetic trace
13
SE results P2P traffic
Captured trace
Synthetic trace
14
Traffic sensitivity with regards to the
distributions

• Random Variables (R.V.) distributions?
• Heavy-tailed distributions important?
• Source of correlation in traffic?
• Investigation of each R.V. separately
• Replace individually the empirical distribution
  of the studied R.V. by a memoryless distribution
• Model the other R.V. by the empirical
  distributions
• Impact on the spectra?
• Conclusion on the importance of the R.V.
  distribution

15
Mail traffic sensitivity
Insensitive distributions
Sensitive distributions
16
P2P traffic sensitivity
Insensitive distributions
Sensitive distributions
17
Conclusion

• Extended LiTGen reproduces accurately the
  traffic scaling properties
• Investigation of the impact of the R.V.
  distributions
• The in-objects organization is crucial
• Heavy-tailed distribution correlation
• Give insights for the development of accurate
  traffic models

18
Future works

• Dependency introduced in Extended LiTGen
• Realistic performance prediction?
• Burstiness strong implications on queuing
  performance
• Compare the performance of a model fed by
• The captured traffic
• The synthetic traffic from LiTGen
• Simpler renewal processes

19
Thank you !
20
Trace originating on the Sprint access network