dgLBF

Declarative implementation of guaranteed Latency Based Forwarding (gLBF).

How To  

Download or clone this repo. Make sure you have the following prerequisites:

• swipl >= 9.0.4
• python >= 3.8
• swipl MQI for Python
• requirements.txt for the python virtual environment

1. Create a virtual environment and install the required packages:

   python3 -m venv venv
   source venv/bin/activate
   pip install -r sim/requirements.txt

2. Then, run the python script in the sim directory, after activating the virtual environment:

   python3 main.py [OPTIONS]

   The script has the following options:

   Usage: main.py [OPTIONS]
   
   Start an experiment with an infrastructure of NODES nodes, and FLOWS flows.
   
   Options:
   -f, --flows INTEGER             Number of flows in the experiment.
                                   [required]
   -n, --nodes INTEGER             Number of nodes in the infrastructure.
   -g, --gml [janos-us-ca|giul39|germany50|ta2|norway|cost266|atlanta|di-yuan|pdh|india35|sun|geant|dfn-bwin|france|polska|dfn-gwin|nobel-eu|abilene|newyork|janos-us|nobel-germany|pioro40|zib54|nobel-us]
                                   Name of a GML file (in data/gml) to use as
                                   infrastructure.
   -i, --max_iterations INTEGER    Number of of trials to find a solution for
                                   each combination #flows / infrastructure.
   -s, --seed INTEGER              Seed for the random number generator.
   -t, --timeout INTEGER           Timeout for the experiment.
   --help                          Show this message and exit.
   

   N.B., using the --gml option will override the --nodes option. Instead, by using the --nodes option, the script will generate a random infrastructure with the specified number of nodes. Flows are generated randomly, with a random source and destination node, and a random latency requirement.

   Change the ranges of random values for both infrastructures and flows in the config.py script.