Quadratic Assignment Problem

Implementation of Quadratic Assignment Problem using Minimum Congestion Algorithm by Bansal et. al.

Currently the code only supports random graph with no user input.
Substrate Graph- Random connected graph with random capacity and weights/costs.
Workload Graph- Star graph with random number of leaf nodes, random uniform edge demand and node demand of 1.

How to use:

1. Create a pytohn virtual environment
2. Run pip install -r requirments.txt
3. Create dataset/internet directory to add graphs for internet topology.
4. Run python algorithm.py [-t/--topology <topology>] [-sg/--save_graph] [-sd/--save_drive]
5. To save all the results to google drive.
   1. Create a project in google cloud.
   2. Create a service account to avoid authorization before run.
   3. Save service account key as client_sercets.json.
   4. Get folder_id from Google drive URL.
   5. Add {"folder_id": $folder_id_fetched_from_url} in config.json.

Note: Currently we support only Internet Topology. Clos, Bcube, and Xpander topologies will be added in next iterations.