Google Professional Data Engineer exam dumps

Google Professional Data Engineer practice question 111 of 279

Professional Data Engineer. Professional level, Google Cloud. Free question with the correct answer and a full explanation.

Google Professional Data Engineer Question 111

Select 3Google Cloud Platform

You are designing a data pipeline in Google Cloud for a retail company. The pipeline ingests real-time customer transaction data using Pub/Sub, processes the data using Dataflow, and stores the output in BigQuery for analytics. As part of operationalizing this pipeline, you need to ensure it is reliable, cost-efficient, and easy to monitor. Which of the following practices should you implement?

  1. A

    Enable Dataflow Auto-scaling to optimize resource usage during fluctuating workloads.

  2. B

    Set up Pub/Sub Dead Letter Topics to handle message delivery failures.

  3. C

    Use manual Dataflow worker node allocation to maintain predictable costs.

  4. D

    Activate BigQuery Slot Reservations to control query execution costs.

  5. E

    Implement Cloud Monitoring alerts to notify the team of pipeline errors or performance issues.

Show answer and explanation

Correct answers: A, B, E

Explanation

Operationalizing a data pipeline involves ensuring reliability, cost efficiency, and effective monitoring. Practices like enabling Dataflow Auto-scaling, setting up Pub/Sub Dead Letter Topics, and implementing Cloud Monitoring alerts directly address these needs. Manual worker node allocation and BigQuery Slot Reservations are less relevant or inefficient for this scenario.

  • A. Correct.

    Enabling Dataflow Auto-scaling dynamically adjusts the number of worker nodes to handle changes in workload, optimizing both cost and performance. This is a recommended practice for operational efficiency.

  • B. Correct.

    Setting up Pub/Sub Dead Letter Topics ensures that undeliverable messages are logged and can be analyzed later, improving reliability and fault tolerance in the pipeline.

  • C. Incorrect.

    Manual Dataflow worker node allocation can lead to underutilization or over-provisioning of resources, making it less efficient than auto-scaling for fluctuating workloads.

  • D. Incorrect.

    Activating BigQuery Slot Reservations is useful for managing costs in environments with predictable, high-volume queries, but it is not directly relevant to operationalizing real-time pipelines.

  • E. Correct.

    Cloud Monitoring alerts help teams quickly identify and address errors or performance issues in the pipeline, improving reliability and maintainability.

Timed practice exam

Take a Google Professional Data Engineer practice test under exam conditions

60 questions in 120 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam