Salesforce Certified Data 360 Consultant free question 4 / 10 · Domain 3: Data Source Connection and Ingestion
You run two batch data transforms, A and B, on an hourly schedule starting at midnight every day. Both use a DLO as input, use only filters and single-row formulas, and each run changes about 2% of the input. A uses the Replace output write mode, and B uses Upsert. Which statement matches the Help article's explanation of run types for the 10:00 AM run on a given day?
- B runs incrementally, and A runs as a full run
- Both A and B run as incremental runs that handle the changed records
- A runs incrementally, and B runs as a full run
- Regardless of write mode, both run as full runs at this time
Show the answer and explanation
Answer: C
Why this is correct
The article says an incremental run needs a scheduled run, input changes under 5 percent of the total, an input that is a DLO or DMO, and the Replace output write mode. With Replace, the transform first tries an incremental run, and for an incremental run it upserts or deletes only the changed records. A data transform that uses the Upsert write mode runs as a full run. The first scheduled run of the day is full for every data transform, but 10:00 AM is not the first run of the day, so A runs incrementally and B runs in full.
Why the other options are wrong
"Both A and B run as incremental runs that handle the changed records" … A runs incrementally, but a data transform that uses the Upsert write mode runs as a full run, so B does not run incrementally.
"B runs incrementally, and A runs as a full run" … This reverses the two. Replace is one of the conditions for an incremental run, and choosing Upsert makes the transform run in full.
"Regardless of write mode, both run as full runs at this time" … B does run in full, but A meets every condition for an incremental run, and it is not the first run of the day, so A runs incrementally.
Other cases that force a full run
A manual run through Run Now or a flow, using aggregates, multi-row formulas, or joins, and changes above 5 percent of the total also cause a full run. After you edit the transform, change the DMO mapping of an input, or recreate an input data object, the next run is full. You cannot choose the run type yourself. The transform decides it from these conditions.