Specialised Environments

Page tree

Versions Compared

Old Version 4

changes.mady.by.user Rahul Mahendru

Saved on

compared with

New Version Current

changes.mady.by.user Rahul Mahendru

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

You could request multiple GPU nodes  as below (2 nodes as in the example form)

Step1:

  • Use 'gpuvolta' in the 'Queue' field
  • Select 'Custom' in the 'Compute Size' pull down menu
  • Request > 4 GPUs
    • 8 GPU devices from 2 nodes in the form below

Step 2:

  • Click on 'Advanced Options' button
  • Use '/g/data/dk92/apps/Modules/modulefiles' as the Module Directory
  • Load either 'rapids/23.06/runtime-py310' or 'rapids/23.06/runtime-py39' and 'gadi_jupyterlab/22.06' modules
  • Put 'jupyterlab.ini.sh -D g' in the Pre Script field
  • Click 'Launch' button

Image Modified

Image Modified

Wait until the "Open JupyterLab" button is highlighted and then click it to open the JupyterLab interface.

...

Now you can see 8 workers come from 2 nodes, each node has 4 workers, each worker occupies a GPU device ( Tesla V100-SXM2-32GB).

Monitor GPU status

Dask-dashboard

You can monitor GPU status via the Dask-Dashboard after setting up the localCUDACluster or connecting to a pre-defined Dask GPU cluster in your notebook.

For example, you click "WORKER" button to view all worker activities as below

Step1:

  • Click on 'Dask' button from left sidebar
  • Select 'WORKERS' metric

Step 2:

  • You will see runtime information of all workers in a new tab

Image Added

Image Added

For more details on Dask dashboard, please refer to here.

You can also use NVDashboard by clicking the “GPU Dashboards” menu along the left-hand side of your JupyterLab environment.

Image Added

For more details on NVDashboard, please refer to here.