Databricks Certified Data Engineer Professional DATA-ENGINEER-PROFESSIONAL — Question 67

A DLT pipeline includes the following streaming tables: • raw_iot ingests raw device measurement data from a heart rate tracking device. • bpm_stats incrementally computes user statistics based on BPM measurements from raw_iot. How can the data engineer configure this pipeline to be able to retain manually deleted or updated records in the raw_iot table, while recomputing the downstream table bpm_stats table when a pipeline update is run?