.. _tutorials-examples-index:

Slurm Examples
==============

The following examples demonstrate common CoSTAR job submission patterns, from a basic Slurm job through containerised Python,
GPU, checkpointing, and MPI workflows.

.. raw:: html

   <div style="margin-bottom: 50px;"></div>

.. grid:: 1 2 2 2
   :gutter: 3
   :margin: 1

   .. grid-item-card:: Example 1: Basic Serial Batch Job
      :link: tutorials-example-01-helloworld
      :link-type: ref

      New to Slurm? Start here. Learn how to submit a basic serial batch job and understand Slurm output files.

   .. grid-item-card:: Example 2: Python/Conda + Apptainer
      :link: tutorials-example-02-python
      :link-type: ref

      Build a small Apptainer image from a public Docker Python image, create a Conda environment inside it, and run a CPU-only Python calculation.

   .. grid-item-card:: Example 3: PyTorch + MNIST + GPU
      :link: tutorials-example-03-gpu
      :link-type: ref

      Train a PyTorch neural network on the MNIST dataset using a GPU and an Apptainer container.

   .. grid-item-card:: Example 4: TensorFlow + CIFAR-10 + checkpoint
      :link: tutorials-example-04-checkpoint
      :link-type: ref

      Train a TensorFlow model on CIFAR-10, save checkpoints after each epoch, and resume from the latest checkpoint when resubmitted.

   .. grid-item-card:: Example 5: Containerized MPI Multi-Process Job
      :link: tutorials-example-05-mpi
      :link-type: ref

      Run a Python mpi4py calculation across multiple Slurm MPI ranks using Apptainer and the host MPI launcher.

.. raw:: html

   <div style="margin-bottom: 50px;"></div>

.. important::
   Please remember **the Golden Rule of High Performance Computing:**

   Never run heavy calculations, compile massive codebases, or train models directly on the login node! Always use Open OnDemand or the SLURM scheduler to request a dedicated compute node.




.. toctree::
   :hidden:

   example-01-helloworld
   example-02-python
   example-03-gpu
   example-04-checkpoint
   example-05-mpi