Jupyter Notebooks

0. Connect to Bigelow's VPN
1. Start an interactive session
For c1 or c3:
qsub -I -q route -l walltime=8:00:00 -l ncpus=2,mem=16G -N jupyter-interactive
Use an additional -l model=<model> option if the user wants to specifically use c1 or c3.
For c2:
qsub -I -V -q scgc-route -l walltime=8:00:00 -l ncpus=2,mem=32G -N jupyter-interactive
2. Navigate to your notebooks directory, set up environment:
Run:
module load anaconda3
​
unset XDG_RUNTIME_DIR
3. Determine the IP address of the compute nodes
Run:
chost
The hostname and ip address be displayed.
4. Start a jupyter notebook session
Run:
jupyter notebook --no-browser --port=<port> --ip=<ip address>
The port can be in the range of 8888-8898. The ip address is the IP found in the previous step. For example, if you are in c1, to start a session, run:
jupyter notebook --no-browser --port=8888 --ip=10.10.1.1
You should get ouput that looks like this:
Copy/paste this URL into your browser when you connect for the first time,
    to login with a token:
        http://10.10.1.1:8888/?token=3258d683aae77486e9880349b0c61b2412f32ae787a9463b
Note the 'port' value in the URL. Above it is '8888', but this could vary if others are also running notebooks on the server you're working on.
5. Copy and paste the URL from your notebook session into a browser on your local machine.
In the above example you would copy "http://10.10.1.1:8888/?token=3258d683aae77486e9880349b0c61b2412f32ae787a9463b", open your browser of choice, and paste it into the navigation bar.

Last modified 7mo ago