Hi Dirk,
The word 'step' here denotes the transformations in BODS. So the requirement is how to control flow of data, so that the next transformation block gets processed only when previous one has been completely processed till the very last record.
Out of 3 suggestions you gave, can you please elaborate the 2nd and 3rd one?
What is in-memory sort in Query transform? Merge cannot be a solution in this regard because it is scenario dependent when we have atleast two datasets.
Can you please elaborate how can we use post-load command in order to implement this?
Your suggestions are full of learning though!
Thanks