This project is a combination of multipath routing (network load balancing) and server load balancing.
Here the functions of each files/folders are explained.
| File path | Description |
|---|---|
| controller.py | Ryu Controller. |
| flask_server.py | Flask app running on few host |
| load_balancer_logic.py | Refreshs server stats and holds logic |
| load_balancer.py | Flask app that does load balancing |
| Makefile | Cleanup previous MiniNet topology |
| MiniNAM.py | Realtime network visualizer |
| rand_req.sh | Queries service from load balancer randomly, every client runs this. |
| run.sh | Runs Falsk server on clients |
| run_server.sh | Runs "run.sh" with memory, CPU constrains. |
| stats.py | Collects stats on the machine it is running on (server). |
| topo_mini.py | Small network for multipath testing. |
| topo_scalar.py | Full scale data center topology. |
| visualizer.py | Visualizing stats collected by load balancer. |
Step 1: Open terminal from the folder location. Run the following command.
ryu-manager controller.py --observe-links
Step 2: Open another terminal and run the command.
sudo python topo_scaler.py
Step 3: In the Mininet CLI run the following command to open terminal on each server and on load balancer.
xterm h1 & xterm h2 & xterm h3 & xterm h100 & xterm h100 & xterm h10 & xterm h4 & xterm h5
Step 4: Execute the following command on h1, h2, h3, h4 and h5. Here resources dedicated are 700 MB is memory constrain and 100% i.e. one core.
./run.sh -m 700 -c 100 -h <the host number>
Step 5: Run load balancer in h100
<path to python executable> load_balancer.py 10.0.0.47
Step 6: Launch and run visualizer in h100
brave-browser --no-sandbox & <path to python executable> visualizer.py
Step 7: When the Brave Browser opens, go to 10.0.0.47:8000 for visuals.
Step 1: Run the Mininet file
sudo python MiniNAM.py --custom topo_mini.py --topo mytopo --controller remote
Step 2: Run Ryu manager.
ryu-manager controller.py --observe-links
-
Brave Browser
-
Installations from requirements.txt
