PyDarshan Documentation

Python utilities to interact with Darshan log records of HPC applications. PyDarshan requires darshan-utils version 3.3 or higher to be installed.


  • Darshan Report Object for common interactive analysis tasks

  • Low-level CFFI bindings for efficient access to darshan log files

  • Plots typically found in the darshan reports (matplotlib)

  • Bundled with darshan-utils while allowing site’s darshan-utils to take precedence


For examples and Jupyter notebooks to get started with PyDarshan make sure to check out the examples subdirectory.

A brief examples showing some of the basic functionality is the following:

import darshan

# Open darshan log
with darshan.DarshanReport('example.darshan', read_all=False) as report:

    # Load some report data
    # or fetch all

    # ...
    # Generate summaries for currently loaded data
    # Note: aggregations are still experimental and have to be activated:


To install in most cases the following will work:

pip install --user darshan

For alternative installation instructions and installation from source refer to <docs/install.rst>


Targets for various tests are included in the makefile. To run the normal test suite use:

make test

Or to test against different version of Python using Tox:

make test-all

Coverage tests can be performed using:

make coverage

Conformance to PEPs can be tested using flake8 via:

make lint


Documentation for the python bindings is generated seperatedly from the darshan-utils C library in the interest of using Sphinx. After installing the developement requirements using pip install -r requirements_dev.txt the documentation can be build using make as follows:

pip install -r requirements_dev.txt
make docs

File List

  • darshan::

    core darshan python module code

  • devel::

    scripts for building python wheel

  • docs::

    markdown documentation used by sphinx to auto-generate HTML RTD style doc

  • examples::

    Jupyter notebooks showing PyDarshan usage with log files

  • tests::

    PyDarshan specific test cases

  • requirements.txt::

    pip requirement file for minimum set of depednencies

  • requirements_dev.txt::

    pip requirement file for depednencies needed to run development tools


    python file for building/generating PyDarshan package

  • setup.cfg::

    input for


    input files for package

  • tox.ini::

    input for tox which runs the automated testing