I now have a distributed database running on some raspis with > 1Billion rows.
If anyone has ideas for example datasets and questions for the cluster I'm all ears.