Parallel Programming with DASK on GPU

SDSC Expanse Notebook: Parallel_Programming

This README file provides instructions for Expanse users to run Parallel_Programming notebooks in the Expanse. Introduces the Dask module with a simple example and illustrates the Dask graph.
Listof Content

Import Module:

  • dask
  • mkl
  • da
  • numpy

Launch Galyleo

For specific information about launching Galyleo, please refer to this GitHub repository.

Environment Modules

This command line loads GPU modules and Anaconda3 in the Jupyter session to run in a GPU environment.

  • GPU: --env-modules gpu/0.17.3b,anaconda3/2021.05
    galyleo launch --account abc123 —partition gpu-shared --cpus 10 --memory 92 --gpus 1 --time-limit 00:30:00  --env-modules  gpu/0.17.3b,anaconda3/2021.05 --bind /oasis,/scratch --nv
    

    Install Modules

    To run Parallel_Programming notebooks, we do not need to install any additional packages.

Location

Parallel_Programming
├── dask_graphs.ipynb
├── README.md

Submit Ticket

If you find anything that needs to be changed, edited, or if you would like to provide feedback or contribute to the notebook, please submit a ticket by contacting us at:

Email: consult@sdsc.edu

We appreciate your input and will review your suggestions promptly!




Enjoy Reading This Article?

Here are some more articles you might like to read next:

  • Parallel Programming with DASK on CPU
  • Hello_World CPU
  • Hello_World GPU
  • Data Analysis with CuPy
  • Law of Cosines on a CUDA GPU (NVIDIA)