Exporting Iris data to various formats is a data-intensive task that requires to fetch and transform 100GB+ of data per measurement. The original approach was to generate all the formats server-side and upload them on S3, however this was very expensive in terms of compute power and storage. The current approach is to export a single format, and to let the user convert it to its desired format with pantrace.
docker run ghcr.io/dioptra-io/iris-exporter:main --helpThe default settings of the exporter match the default settings of Iris. As such, no particular configuration is required to run the exporter against a local instance of Iris.
# In the iris repository:
docker compose up -d
docker compose exec api .venv/bin/alembic upgrade head
# Ensure that the `public-exports` bucket exists:
# https://docs.min.io/docs/minio-client-quickstart-guide.html
mc alias set docker http://minio.docker.localhost minioadmin minioadmin
mc mb docker/public-exports
# In the iris-exporter repository:
python -m iris_exporter TAG