This tutorial focuses on managing Python environments with Conda and using them effectively inside Jupyter Notebooks on Hyak via Open OnDemand.
💡 TIP: While examples are demonstrated on Klone, the same workflow applies to Tillicum with minor modifications (e.g., using QOS instead of partition for Slurm).
Working with Conda in an HPC environment can feel confusing at first — especially when combining:
- Lmod software modules
- Conda environments
- Slurm job scheduling
- Jupyter via Open OnDemand
This tutorial is designed to connect all of these pieces into one coherent workflow that you can reuse for your research projects.
🎯 By completing this tutorial, you’ll learn how to:
- Load software using LMOD modules
- Create and manage isolated Conda environments
- Register Conda environments as Jupyter kernels
- Launch and run Jupyter through Open OnDemand
- Run Python scripts interactively and via Slurm batch jobs
Each topic in this tutorial is contained in its own Markdown file for easy navigation:
| Section | Description |
|---|---|
| 00-preparation.md | Account and access prerequisites |
| 01-modules.md | Using Lmod to load software modules |
| 02-conda-env.md | Creating, activating, and managing Conda environments on Hyak |
| 03-python.md | Running Python scripts via Slurm (interactive and batch) |
| 04-ood-jupyter.md | Launching Jupyter via Open OnDemand and switching kernels |
| 05-task.md | Hands-on exercise: build an environment and use it in Jupyter |
An optional introduction video accompanies this tutorial and provides a high-level walkthrough of the concepts covered:
March 5, 2026 Managing Python Environments with Conda in Jupyter on Hyak Workshop will be added here when available.
Slide Deck from live tutorial on March 5, 2026
We’d love your feedback to help improve this tutorial and future Hyak trainings. After completing the tutorial or attending the workshop, please take a moment to fill out our feedback form.
🗓️ Stay tuned: Check our Research Computing Calendar for upcoming training events.