Thicket Tutorial

Thicket is a python-based toolkit for Exploratory Data Analysis (EDA) of parallel performance data that enables performance optimization and understanding of applications' performance on supercomputers. It bridges the performance tool gap between being able to consider only a single instance of a simulation run (e.g., single platform, single measurement tool, or single scale) and finding actionable insights in multi-dimensional, multi-scale, multi-architecture, and multi-tool performance datasets.

You can find detailed documentation, along with non-interactive tutorials from this repository of Thicket in the ReadTheDocs.

This repository contains materials for Thicket's hands-on tutorial. You can do all of the exercises on your own laptop using BinderHub.

You find Thicket itself in its Github repository: https://github.com/llnl/thicket

Running the notebooks online

We use BinderHub to create a shareable and interactive environment of the notebooks within a live JupyterHub instance.

You can access the interactive environment at this link or by clicking the badge at the top of this file.

Running the notebooks locally

Docker

We provide a Dockerfile for users to run the notebooks locally. To run locally and interactively, you must first build the Docker container with:

$ docker build -t thicket-tutorial -f Dockerfile .

Then, you must create a Docker network with:

$ docker network create jupyterhub

Finally, you can launch the tutorial. To launch the tutorial without preserving any changes, run:

$ docker run --rm -it --entrypoint /start.sh -v /var/run/docker.sock:/var/run/docker.sock --net jupyterhub --name jupyterhub -p 8888:8888 thicket-tutorial

If you would rather your changes be preserved, run:

$ docker run --rm -it --entrypoint /start.sh -v /var/run/docker.sock:/var/run/docker.sock -v .:/home/jovyan --net jupyterhub --name jupyterhub -p 8888:8888 thicket-tutorial

Alternatively, if you want to run the notebooks automatically (i.e., non-interactive), you can simply run the dev_scripts/autorun.sh script. This script executes the same commands as above, but it uses the run_all.sh script as an entrypoint instead of start.sh.

The Docker-based code for running this tutorial locally was derived from the material from the 2023 RADIUSS tutorial for Flux, which can be found here: https://github.com/flux-framework/Tutorials/tree/master/2023-RADIUSS-AWS/JupyterNotebook

Podman

If you want to use podman instead of docker, you can replace "docker" with "podman" for most steps.

First initialize and start podman:

$ podman machine init
$ podman machine start

Then build the container:

$ podman build -t thicket-tutorial -f Dockerfile .

Then create a network:

$ podman network create jupyterhub

Then launch the tutorial:

$ podman run --rm -it --entrypoint /start.sh -v /var/run/docker.sock:/var/run/docker.sock --net jupyterhub --name jupyterhub -p 8888:8888 thicket-tutorial

Clean up after you are done:

$ podman machine stop

License

This repository is distributed under the terms of the MIT license.

All contributions must be made under the MIT license. Copyrights are retained by contributors. No copyright assignment is required to contribute to this project.

See LICENSE.

SPDX-License-Identifier: MIT

LLNL-CODE-834749