Running benchmarks

Setup

Link the SDQL repository

Create a symlink to the main SDQL repository:

# clone the sdql repo somewhere (e.g. home directory)
cd ~ && git clone https://github.com/edin-dal/sdql/
# create a symlink here to the SDQL programs directory
cd - && ln -s ~/sdql/progs progs

Install Python version and virtualenv

Install pyenv and pyenv-virtualenv:

curl https://pyenv.run | bash
# follow these instructions to install pyenv-virtualenv
# https://github.com/pyenv/pyenv-virtualenv?tab=readme-ov-file#installing-as-a-pyenv-plugin

On macOS:

brew install pyenv
brew install pyenv-virtualenv
# follow these instructions if you get "Failed to activate virtualenv"
# https://github.com/pyenv/pyenv-virtualenv/issues/387

Get the required Python version:

export VERSION=$(cat pyproject.toml| grep -Poi '^python = "\K(\d+.\d+.\d+)')

On macOS run it as ggrep after brew install grep to use GNU grep:

export VERSION=$(cat pyproject.toml| ggrep -Poi '^python = "\K(\d+.\d+.\d+)')

Create an environment with the required version:

pyenv install $VERSION
pyenv virtualenv $VERSION sdql

Setting it as the local environment will activate it:

pyenv local sdql

Install Poetry manager and packages

Install the package manager poetry:

curl -sSL https://install.python-poetry.org | python3 -

Install the required packages to your environment:

poetry install

Run

Make sure you have datasets in progs/../datasets/tpch/.

We suggest generating them with s -1.

Then run the benchmarks as follows:

export PYTHONPATH=. && python __main__.py