I have an unpartitioned dataset, I would like to create a new version that is partitioned by a given column value. so it column X has values a,b,c and I would like those to be the 3 partitions. How would I do that?
you can :
- make a partitioned dataset with the column X as discrete partitioning dimension
- do a Sync recipe from your unpartitioned dataset to the partitioned one, and check the "redispatch" checkbox
- put any value for partition dimension and run the recipe