A data engineer is responsible for a Lakeflow job that runs every hour. The job failed during the last run due to a transient network error. The engineer wants to rerun the failed job without waiting for the next scheduled run. What should they do?

Use the 'Restart' function on the job to re-run the entire job from the beginning.

Use the 'Repair' function on the failed run to re-run only the failed tasks.

Create a new job with the same configuration and run it manually.

Use the 'Stop' function on the job and then start it again.

A company uses Azure Databricks for ETL workloads. They want to be notified when a job fails so they can respond quickly. They have already configured diagnostic settings to stream job logs to Log Analytics. What should they do to receive proactive notifications?

Enable the 'Send email on failure' option in the cluster configuration.

Set up a webhook in the Lakeflow job to call an external endpoint on failure.

Create an alert rule in Azure Monitor that triggers on job failure logs.

Use the Databricks CLI to monitor job status and send emails.

A data engineer notices that a cluster is consistently underutilized and wants to free up resources while retaining the cluster configuration for future use. Which action should the engineer take?

Resize the cluster to zero workers

A company wants to centralize monitoring of all Azure Databricks workloads. They need to track job failures, cluster start/stop events, and notebook execution details in a single dashboard. The operations team prefers to use Azure Monitor for alerting. What should you configure?

Enable diagnostic settings on the Databricks workspace to stream logs to a Log Analytics workspace, then build queries and alerts.

Create a Databricks job that runs every hour and writes cluster metrics to a Delta table.

Install the Databricks Monitoring Agent on each cluster node to forward logs to Azure Monitor.

Use the Databricks REST API to pull logs and store them in Azure Blob Storage, then use Azure Data Explorer for analysis.

A data engineering team is tuning a Spark job that performs a large join between two tables. The job runs on a cluster with 16 executor cores. The Spark UI shows that the join stage has 200 tasks, but only 16 tasks run at a time. The team wants to reduce the runtime. They have already tried increasing the number of shuffle partitions, but the runtime did not improve. What should they do next?

Use a broadcast join hint to avoid shuffling.

Increase the parallelism of the cluster by adding more executor cores.

Enable speculative execution to handle slow tasks.

Increase the executor memory to reduce spilling.

Microsoft Certified:Azure Databricks Data Engineer Associate

Domain 4Objective 4

Monitor, Troubleshoot, and Optimize Workloads in Azure Databricks DP-750 Practice Questions (Page 6)

Part of the Deploy and maintain data pipelines and workloads domain, which accounts for 30–35% of the DP-750 exam. Microsoft does not publish an official question count, but from its 120-minute exam (~50–80 total, ~15–28 in this domain), expect 4–7 from this objective — we provide 47 practice questions to prepare you well beyond it. (estimate)

47questions here

10free pages

11concepts

30–35%of the exam

Questions 26–30

26application · medium
A data engineer is responsible for a Lakeflow job that runs every hour. The job failed during the last run due to a transient network error. The engineer wants to rerun the failed job without waiting for the next scheduled run. What should they do?
Select an answer first
27application · medium
A company uses Azure Databricks for ETL workloads. They want to be notified when a job fails so they can respond quickly. They have already configured diagnostic settings to stream job logs to Log Analytics. What should they do to receive proactive notifications?
Select an answer first
28foundation · easy
A data engineer notices that a cluster is consistently underutilized and wants to free up resources while retaining the cluster configuration for future use. Which action should the engineer take?
Select an answer first
29application · medium
A company wants to centralize monitoring of all Azure Databricks workloads. They need to track job failures, cluster start/stop events, and notebook execution details in a single dashboard. The operations team prefers to use Azure Monitor for alerting. What should you configure?
Select an answer first
30expert · hard
A data engineering team is tuning a Spark job that performs a large join between two tables. The job runs on a cluster with 16 executor cores. The Spark UI shows that the join stage has 200 tasks, but only 16 tasks run at a time. The team wants to reduce the runtime. They have already tried increasing the number of shuffle partitions, but the runtime did not improve. What should they do next?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to DP-750

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Microsoft. “DP-750” is a trademark of its owner, used for identification only.