Incremental refresh for a Fabric OneLake data model fails due to resource constraints. Which statement is correct?

• A. Query folding is occurring.
• B. Only refresh complete days is selected.
• C. XMLA Endpoint is set to Read Only.
• D. Query folding is NOT occurring.

Answer: (D)

Incremental refresh works best when the data source can handle as much of the filtering and transformation as possible. This is what query folding does: it pushes the transformations back to the data source so only the new or changed data is loaded and processed. When query folding is not happening, the engine has to pull larger portions of data and perform more work locally, which increases memory and compute needs. Under resource constraints, that extra load can cause the incremental refresh to fail because there isn’t enough capacity to complete the operation.

If folding were in effect, incremental refresh would be more efficient and less likely to hit resource limits, since the data source would do the heavy lifting for the partitions that encompass only the new data. The other options describe scenarios that wouldn’t directly explain a failure caused by running out of resources: limiting refresh to complete days reduces data but doesn’t explain the resource issue; a read-only XMLA endpoint would block the refresh entirely rather than fail specifically for resource constraints.