DP-900 Practice Question: Identify considerations for relational data on Azure

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✓

  Create the target table, then define mapping, then execute the ingestion, then verify.

  Why this is correct

  The target table must exist first because an Azure Data Explorer ingestion mapping references the target table's schema — including column names, data types, and whether columns are nullable or keep-always. After the table is created, you define an ingestion mapping (e.g., csv, json, avro) that tells the ingestion engine how to translate source fields into table columns. Only then can you execute the ingestion command or pipeline, and verification is the final step to confirm rows were committed and to check for any partial failures.

• ✗

  Define mapping, then create the target table, then execute the ingestion, then verify.

  Why it's wrong here

  Defining the mapping before creating the target table is invalid in Azure Data Explorer because a mapping is metadata attached to a specific table — it points to that table's schema for column resolution. If the table does not exist, Kusto will reject the mapping definition command with an error such as 'Table '...' does not exist.' Even when the table is later created, the pre-defined mapping would have been created against nothing and would not correctly associate with the table. The correct sequence is table first, then mapping, then ingestion, then verification.

• ✗

  Execute the ingestion, then verify, then create the target table, then define mapping.

  Why it's wrong here

  Executing ingestion as the first step is impossible because the ingestion engine requires both an existing target table and a defined mapping to know where to route data and how to interpret it. Verification before table creation or mapping makes no practical sense — there is no schema to verify against and no data committed yet. Furthermore, defining the mapping at the very end is useless because mapping is consumed during the ingestion operation, not after; retroactively adding a mapping has no effect on already-ingested data. This order completely inverts the dependency chain of the ingestion pipeline.

• ✗

  Create the target table, then execute the ingestion, then verify, then define mapping.

  Why it's wrong here

  In this ordering, ingestion is attempted without a mapping, but Azure Data Explorer requires a mapping (or at least a valid format-based default) to correctly translate the incoming source fields into the table's columns — without explicit mapping, ingestion can fail with schema-mismatch or leave data incorrectly parsed. Defining the mapping only after ingestion completes is too late, as the ingestion engine has already attempted to process the data using either no mapping or a built-in default that likely does not match the table schema. While creating the table first is correct, skipping the mapping step before execution introduces a fatal gap in the pipeline, and then trying to patch it afterward cannot repair data that was never ingested correctly.

Go deeper

Related to this question