I have a table with +10 million records looking like this:
Then, using PgSQL CUBE I wish to find all the best matches (distinct by recipe_id), with a maximum dist
