About these questions
These five questions are original. We wrote them in the scenario style of the DP-750 skills measured, spread over all four domains. They are not real exam questions, and the real exam will word things differently.
How to use these
Try each question without notes, and choose an answer before you open the explanation. Then read the explanation even when you were right, because the reason the closest wrong option fails is where the learning is. If you slipped in a domain, revisit it on the domains explained page.
Question 1 (Set up and configure an Azure Databricks environment)
A regional grocery chain has a dashboard that runs the same heavy aggregation over daily store sales about fifty times a day. The managers accept figures that are a few hours old. The data team wants faster dashboard loads and lower repeated compute cost. Which object should back the dashboard?
- A. A standard view that recomputes the aggregation for every query
- B. A dynamic view that filters rows for each manager
- C. A temporary view created when each dashboard session starts
- D. A materialized view refreshed on a schedule
Show the answer and explanation
Correct answer: D. A materialized view stores the precomputed result and refreshes it on a schedule, so repeated dashboard queries read stored data instead of rerunning the aggregation. Slightly stale data is acceptable here, which makes that trade-off a good one. A standard view is the closest wrong option: it is always current, but it reruns the expensive query each time, which is exactly the cost the team wants to avoid. A dynamic view adds per-user filtering but still computes at query time, and a temporary view stores nothing at all.
Question 2 (Secure and govern Unity Catalog objects)
A finance team group already has USE CATALOG on the catalog `finance`. A data owner grants the group USE SCHEMA and SELECT on the schema `finance.invoices`. Two weeks later, an engineer creates a new table in that schema. What access does the group have to the new table?
- A. SELECT, because the schema-level grant is inherited by tables created later
- B. No access, because only catalog-level grants are inherited
- C. No access until the owner grants SELECT on the new table
- D. SELECT on the new table and on tables in every other schema of the catalog
Show the answer and explanation
Correct answer: A. Unity Catalog privileges inherit downward. SELECT on a schema, together with USE SCHEMA and USE CATALOG, covers every current and future table and view in that schema. The closest wrong option is C, which sounds careful but describes a model with no inheritance. Option B wrongly limits inheritance to the catalog level. Option D goes too far, because a grant on one schema never reaches other schemas.
Question 3 (Prepare and process data)
A logistics company is moving 25 TB of historical Parquet files from a retired storage account into a Unity Catalog Delta table. The load runs once, and no new files will ever land in that path. The team wants the simplest setup, with no streaming query and no checkpoint to manage. Which approach fits best?
- A. An Auto Loader stream with schema and checkpoint locations
- B. A COPY INTO command that loads the files into the Delta table
- C. A Structured Streaming query with an hourly processing-time trigger
- D. A query-based Lakeflow Connect connector pointed at the Parquet path
Show the answer and explanation
Correct answer: B. COPY INTO is a batch SQL command that loads files into a Delta table with little configuration, which suits a one-time load of a large file set. Auto Loader is the closest wrong option. It shines for continuous, incremental arrival of files, but for a single load its streaming model and checkpoint management add complexity you do not need. A processing-time trigger keeps a stream running long after the load ends. Query-based connectors read relational tables, not files in cloud storage.
Question 4 (Deploy and maintain data pipelines and workloads)
An hourly Lakeflow Job loads shipment records and sometimes runs for 80 minutes. Two runs must never write to the same tables at once, and the business requires that no scheduled run is ever lost, even when a previous run overruns. Which two settings should you configure? Choose two.
- A. Set maximum concurrent runs to 2
- B. Set maximum concurrent runs to 1
- C. Queue runs that exceed the limit
- D. Skip runs that exceed the limit
- E. Cancel runs that exceed the limit
Show the answer and explanation
Correct answers: B and C. A limit of one concurrent run prevents two runs from writing to the same tables at the same time. Queuing makes a trigger that arrives during an overrun wait for a free slot and then run, so no scheduled execution is lost. Skip is the closest wrong option because it also avoids overlap, but it drops the waiting run, which the business forbids. Cancel discards the new run too, and allowing two concurrent runs brings back the overlap you must avoid.
Question 5 (Deploy and maintain data pipelines and workloads)
A job uses a table update trigger. During a planned three-hour storage maintenance window, no new runs may start. The run already in progress must be allowed to finish, and the trigger configuration must be kept. What should you do?
- A. Pause the trigger, then resume it after the window
- B. Delete the trigger and add it again afterward
- C. Cancel the active run from the Runs tab
- D. Delete the job and redeploy it after the window
Show the answer and explanation
Correct answer: A. Pausing a trigger stops new runs from starting while a run already in progress is allowed to complete, and resuming restores the same configuration. Deleting the trigger, option B, also stops new runs, but it throws away settings such as the monitored tables, which you need to keep. Cancelling the active run stops work that should finish and does nothing about future runs. Deleting the whole job loses run history and adds needless work.
More guides for DP-750
- DP-750 Exam Format, Scoring and Retakes
- DP-750 Study Plan: 4-Week and 8-Week
- DP-750 Prerequisites and Skills Check
- DP-750 Renewal and Retakes Explained
- Is DP-750 Worth It? An Honest Look
- DP-750 Cheat Sheet: Key Facts to Review
- DP-750 vs DP-700: Which Exam First?
- DP-750 vs Databricks Data Engineer Pro
Next step
Want more? The DP-750 practice tests have five full sets. To judge your readiness, read Is DP-750 hard?.