GeoBench

GeoBench is a tool designed to benchmark geospatial software. It allows users to run benchmarks based on scenarios specified in YAML files or directly in Jupyter notebooks. This project is currently under development.

Prerequisites

Before you begin, ensure you have the following installed:

• Operating System: Linux, Windows (with WSL2 or Powershell), MacOS
• Python: Python 3 with the pip package manager installed

Preparation

Follow these steps to set up GeoBench:

• Clone the repository:

  git clone https://github.com/ITC-CRIB/geobench
  cd geobench

• Pull the repository:

  cd geobench
  git pull

Installation as Tool

• Using pip

  pip install .
  

• Using uv
  • Install uv tool
    • MacOS and Linux

     curl -LsSf https://astral.sh/uv/install.sh | sh
    

    • Windows (from Powershell/Terminal)

     powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
    

  • Run the installation process

   uv tool install . --upgrade
  

• To run the program you can execute following command:

   geobench scenario.yaml
  

Development

• Set up and activate the Python environment:

  • Using venv:

    python3 -m venv .venv
    source .venv/bin/activate   # On Windows, use `.venv\Scripts\activate`
    pip install -r requirements.txt

  • **Using uv (recommended) **:

   uv sync
  

• To run the program during development (without installing it as tool), you can execute following command:

   python -m geobench.cli scenario.yaml
  

Running Benchmarking Scenario

A benchmarking scenario is represented in a YAML file. For the time being, there are two scenario types supported in this tool: qgis-process and qgis-python. The qgis-json and arcgis-command will be supported later.

Using Jupyter Notebook Benchmarking

In addition to YAML scenarios, GeoBench now supports benchmarking directly within Jupyter notebooks. This is useful for benchmarking Python code interactively.

Feel free to contribute to the development of GeoBench by submitting issues or pull requests on the repository. For more details on contributing, please refer to the CONTRIBUTING.md file in the repository.

Jupyter Notebook Benchmarking

GeoBench now supports benchmarking directly within Jupyter notebooks. There are three ways to use this functionality:

1. Using the Geobench Class

from geobench import Geobench

# Create a benchmark instance
bench = Geobench(
    name="my-benchmark",
    outdir="results",
    run_monitor=2.0,
    clean=True
)

# Start benchmarking
bench.start("my-function")

# Run your code
result = my_function()

# Finish benchmarking
bench.stop(True)  # Pass True for success, False for failure

# Generate HTML report
bench.generate_report()

2. Using the @geobench Decorator

from geobench import geobench

@geobench(name="my-function-benchmark", outdir="results", clean=True)
def my_function():
    # Your code here
    return result
    
# Call the decorated function
result = my_function()