Example Notebooks

These Jupyter notebooks demonstrate the key features of iops-profiler with practical, hands-on examples.

Notebooks

Running the Notebooks

You can run these notebooks in several ways:

  1. View Online: The notebooks are rendered as part of this documentation

  2. Download and Run Locally:

    >> git clone https://github.com/lincc-frameworks/iops-profiler.git
>> cd iops-profiler/docs/notebooks
>> jupyter notebook

Prerequisites

Before running the notebooks, make sure you have:

  • Python 3.10 or higher

  • Jupyter Notebook or JupyterLab installed

  • iops-profiler installed (pip install iops-profiler)

The notebooks will handle creating and cleaning up temporary files automatically.

Notebook Descriptions

Basic Usage

Basic Usage

Learn the fundamentals of iops-profiler:

  • Loading the extension

  • Using line magic (%iops) for single-line profiling

  • Using cell magic (%%iops) for multi-line profiling

  • Understanding the results table

  • Comparing different I/O strategies

  • Working with text and binary files

Recommended for: First-time users and those wanting a quick introduction.

Histogram Visualization

Histogram Visualization

Explore the histogram feature for visualizing I/O patterns:

  • Enabling histogram mode with --histogram

  • Understanding operation count and bytes distributions

  • Analyzing read vs. write patterns

  • Optimizing buffer sizes based on histograms

  • Working with mixed operation sizes

  • Real-world examples (CSV files, etc.)

Recommended for: Users who want to understand and optimize I/O operation distributions.

Note: Histogram mode is available on Linux and macOS, but not Windows.

Additional Resources

After working through the notebooks, check out:

You can also find more examples in the project’s GitHub repository.