Traffic Modeling Project

Overview

In this project, our goal is to create models of application network traffic for the purpose of realistic traffic generation.

In the project we have designed and implemented a framework for traffic modeling and generation. This frame work includes many software tools for dataset extraction, traffic modeling, traffic generation, and for evaluating the performance of traffic modeling and generation algorithms.

The components of the framework are designed to be easily extensible to enable addition of new traffic modeling algorithms, and their corresponding traffic generation methods.

Software Applications

A list of software developed as part of the framework, along with their documentation and source code, is provided below:

  • trafficdatasetmaker:
    extracts a dataset with many fields that are useful for realistic traffic
    modeling from raw pcap traces.
    documentation
    source code
  • trafficmodelchecker:
    simulates the generation of traffic protocol data units (PDUs) on a
    network by using a traffic model.
    documentation
    source code
  • traffic_metrics:
    analyzes multiple in network trace files, comparing them based on the
    distribution of metrics calculated from each input trace. It is used
    to evaluate the performance of traffic models by comparing real traffic
    with model-generated traffic.
    traffic_metrics documentation
    source code

Traffic Generator Survey

Network traffic workloads are widely utilized in applied research to verify correctness and to measure the impact of novel algorithms, protocols, and network functions. We conducted a comprehensive survey of traffic generators referenced by researchers over the last 13 years, providing in-depth classification of the functional behaviors of the most frequently cited generators. We developed a tool to analyze about 7000 papers to identify popular traffic generators utilized in research.