Hatchet is a Python-based library that allows Pandas dataframes to be indexed by structured tree and graph data. It is intended for analyzing performance data that has a hierarchy (for example, serial or parallel profiles that represent calling context trees, call graphs, nested regions’ timers, etc.). Hatchet implements various operations to analyze a single hierarchical data set or compare multiple data sets, and its API facilitates analyzing such data programmatically.
To use hatchet, install it with pip:
$ pip install llnl-hatchet
Or, if you want to develop with this repo directly, run the install script from
the root directory, which will build the cython modules and add the cloned
directory to your PYTHONPATH:
$ source install.sh
See the Getting Started page for basic examples and usage. Full documentation is available in the User Guide.
Examples of performance analysis using hatchet are available here.
The hatchet tutorial provides a step by step introduction to hatchet features through a live interactive environment in BinderHub.
Hatchet now provides an interactive visualization which can be run inside of your Jupyter notebooks. It is dependant on different mechanisms for building which we describe here:
The software in the hatchet/vis subdirectory requires Node.js and Node Package Manager for the development and building of javascript code.
One you have Node and NPM installed on your system, you can install all necessary node packages by running the following line in your terminal from the hatchet/vis/ directory:
npm install
To build out javascript into the static bundles used by the Jupyter visualizations, run the following line from the hatchet/vis directory in your terminal:
npm run build
Alterntinatively if you are developing a visualization, you can run the following line to force bundles to automatically update when you change the javascript source code:
npm run watch
Hatchet is an open source project. We welcome contributions via pull requests, and questions, feature requests, or bug reports via issues.
Many thanks go to Hatchet's contributors.
If you are referencing Hatchet in a publication, please cite the following paper:
- Abhinav Bhatele, Stephanie Brink, and Todd Gamblin. Hatchet: Pruning the Overgrowth in Parallel Profiles. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC '19). ACM, New York, NY, USA. DOI
Hatchet is distributed under the terms of the MIT license.
All contributions must be made under the MIT license. Copyrights in the Hatchet project are retained by contributors. No copyright assignment is required to contribute to Hatchet.
See LICENSE and NOTICE for details.
SPDX-License-Identifier: MIT
LLNL-CODE-741008