CSCI-SHU 308 Notes

1. How to set up the environment

Prerequisites:

• Python3. Please note some old versioned macOS and Linux defaults to Python 2.
• pip (shipped with Python in most cases)
• networkx pip package

pip3 install networkx

2. How to use pre-generated datasets

There are three pre-generated network flow data for a same 25-nodes topology called "ATT_NA (ATT North America)".

directory	#nodes	#edges	#prefixes	Flow count
examples/att_na/AttMpls-10-egress_100	25	57	100	2585
examples/att_na/AttMpls-10-egress_1000	25	57	1000	25456
examples/att_na/AttMpls-10-egress_10000	25	57	10000	255003

You may start work on load_example.py to see how the flow database is organized. Now this script simply counts the number of flows and randomly print one flow:

$ git clone https://github.com/cytvictor/net2text-CSCI-SHU-308-Fall23
$ cd net2text-CSCI-SHU-308-Fall23/
$ python load_example.py examples/att_na/AttMpls-10-egress_100
Successfully read the example files.
There is a total of 2585 flows in a topology with 25 nodes and 57 edges.

This is a random flow: Flow(path=24 -> 12 -> 13 -> 5 -> 7, organization=Lao Telecommunication IX, prefix=115.84.82.0/24, bandwidth=204.43353242530557, sp=True

This example includes the features mentioned in the paper: path, organization, bandwidth, is-shortest-path.

Appendix 1. Dataset generation

For your reference, they are generated using the following command:

$ cd examples/generator/
$ python example_generator.py ../att_na/AttMpls.graphml ../att_na/AttMpls-10-egress -a

For further explanation on this data generation tool, please find README from Net2Text original repository.