JupyterHub Environment

JupyterHub Environment#

Build and Test Your OpenSees Scripts with Flexibility and Speed

The JupyterHub Environment on DesignSafe runs on a Kubernetes-managed cluster at the Texas Advanced Computing Center (TACC). It provides a highly accessible, interactive platform for developing, testing, and refining OpenSees scripts—whether you’re using Tcl, Python (OpenSeesPy), or both.

This environment is the starting point (and end point) for most workflows. It allows you to iteratively build and test your input scripts in real time before scaling up to HPC batch jobs.

Why Use Jupyter Hub?#

  • Immediate startup — no job queues or allocation requests

  • Full control and feedback — ideal for debugging, exploration, and visualization

  • Integrated tools — notebook interface, terminal, file manager, code editor, and Python console

  • Supports OpenSees, OpenSeesMP/SP, and OpenSeesPy, plus other languages (MATLAB, Julia, R)

  • Direct connection to Tapis — submit HPC jobs from within the notebook

Performance and Resources#

  • Jupyter Hub supports Python, Julia, MATLAB, and R — useful for researchers integrating multi-language workflows.

  • Each Jupyter session runs in its own container, orchestrated by Kubernetes. The system guarantees:

  • Up to 8 CPU cores and 20 GB of RAM per user session

  • Exclusive use of these resources within your container (not shared with other users)

  • Shared physical nodes underneath, which may cause minor I/O contention under heavy load

This environment is optimized for:

  • Interactive model development

  • Small to medium-sized simulations

  • Pre- and post-processing of data

  • Batch job submission to HPC systems

Ways to Run OpenSees Inside Jupyter Hub#

You can run OpenSees in a variety of flexible ways from within this environment:

  1. Run OpenSees at the Terminal

    Launch any Tcl or Python OpenSees script interactively using the terminal.

    → This is your recommended starting point if it is a natural transition from your personal desktop.

  2. Run OpenSeesPy in a Jupyter Notebook

    Embed scripts and results in a single document for reproducibility and sharing.

  3. Run Tcl or Python OpenSees via os.system()

    Execute command-line jobs from within a Python notebook. Works for both sequential and parallel jobs.

    • Note: %run can also work, but doesn’t accept variable substitution.

  4. Run OpenSeesPy in the Python Console

    Useful for quick one-liners, loading external modules, or running python scripts you had developed on your desktop in Spyder or another IDE.

  5. Pre- and Post-Process Your Models

    Use Python, NumPy, Pandas, Plotly, and other libraries to prepare inputs or analyze results.

  6. Submit HPC Jobs Using Tapis

    Automate job submission from notebooks using the Tapis Python SDK (Tapipy) or helper functions.

  7. Use the Integrated File Editor and Manager

    Quickly edit your input files or scripts and organize your workspace.

  8. Upload and Download Files with the Uploader Tool

    Manage your working directory and sync files from external sources.

Summary#

The DesignSafe Jupyter Hub is a powerful entry point into OpenSees workflows. It offers dedicated computing resources in an easy-to-use environment where you can:

  • Develop, debug, and run OpenSees scripts in real time

  • Automate job submission to HPC systems like Stampede3

  • Perform pre- and post-processing without leaving the notebook interface

Use it as your launchpad for scalable computation: once your model is ready, move to HPC for large-scale runs without leaving Jupyter.

Contents