Forum Discussion
Helper VGeneral question on refresh cascade in linked Power BI Dataflows
Setup: I have a core Dataflow which ingests a dozen Azure SQL DB tables. ETL processing is light because we did most of the heavy lifting in T-SQL. Most of the content is dimensional or small fact tables.
The heavier and more volatile fact tables are included in a separate Dataflow, which included linked flows from the core described above.
It looks like a refresh of the core dataflow also triggers a refresh of the dependent workflow. This makes sense with the linked enetities, since they are just pointers back to the core. The refresh presumably takes care of the non-linked entities, i.e. the larger and more volatile fact tables in the dependent workflow.
For daily processing, we only need frequent refreshes on the dependent (transactional) data sources. Static core content only needs daily at best.
What is the recommendation for scheduled refreshes on dependent workflows? Give we are in a Premium capacity, the answer is probably "as often as needed", but there may be other considerations, perhaps when we build-out a third level of the Dataflow chain.
4 Replies
- tommy_g
But I do want to ensure the dependent workflow is refreshed.
Your reply implies that the parent linked workflow does, indeed, cascade down to the dependent child.
By implication, one could establish a top-level refresh schedule (parent) on a frequency as high as the any one of the dependent workflows would need, based on incoming facts throughout the day.
It might be overkill, but also the simplest and most reliable way to attain a consistent reporting base.
I think we are 90% to "accept as solution". How does my theory resonate with you?
