I created a dataset by joining two existing datasets. How would dataset refresh work for all three datasets? Is it required to syncronize the timing of the sync of parent and child datasets?
Hi! If you are joining two existing data sets into a third data set (assuming at least one of them is using SPICE), the result will always be brought into a third SPICE data set and then you need to add your own refresh schedule to it (it is not synced with the original 2 datasets/schedules). If you are just using one SPICE data set in your child data set (and maybe adding more calculated columns to it) then you can optionally set the child to Direct Query and it will be synced with the original parent data set - no need to make an additional schedule for it.
Thanks, Jesse. One more follow-up question. What if I create a dataset [B] with a data-source tables and also join an existing dataset[A]? Will the scheduled refresh for this dataset [B] trigger refresh of both datasource and dataset[A]?
If data set A is using SPICE, then if you join more tables to it from an existing data source, the result will be a new data set B which will be in SPICE and will need to have it’s own schedule configured on it. It will not be synced to A’s schedule.
Thanks! Yes, I am referring to creating and refreshing data sets in SPICE. Looks like for my use-case will require scheduling refresh of A before the refresh schedule of B. Else if B is refreshed first and then A is refreshed after, B will always have data older data of A.