Google Professional Data Engineer Question 54
Select 3Google Cloud PlatformYour organization has an on-premises data warehouse that is no longer meeting business requirements due to its limited scalability and high maintenance costs. Stakeholders have requested that the data warehouse be migrated to a Google Cloud solution to enable real-time analytics and reduce operational overhead. After discussing with stakeholders, you identify the following needs: supporting ad-hoc query workloads, real-time data ingestion, and cost optimization. What steps should you include in your plan to address the desired state?
- A
Migrate the data warehouse to BigQuery and configure BigQuery slots to optimize query performance.
- B
Set up Pub/Sub and Dataflow for real-time data ingestion into BigQuery.
- C
Implement Cloud Spanner for real-time analytics and cost optimization.
- D
Use Data Catalog to enable metadata management and improve data discoverability.
- E
Use a Lift-and-Shift approach to migrate the data warehouse to Compute Engine VMs.
Show answer and explanation
Correct answers: A, B, D
Explanation
To meet the desired state, the plan must include migrating the data warehouse to BigQuery for scalability and ad-hoc query capabilities, setting up real-time ingestion with Pub/Sub and Dataflow, and leveraging Data Catalog for metadata management. These solutions align with the stakeholder requirements for real-time analytics, cost optimization, and improved data access. Cloud Spanner and Lift-and-Shift approaches do not meet the specific needs outlined by the stakeholders.
- A. Correct.
Correct: BigQuery is a scalable, serverless data warehouse solution on Google Cloud that supports ad-hoc queries and cost-effective execution. Configuring BigQuery slots helps optimize query performance.
- B. Correct.
Correct: Pub/Sub and Dataflow are ideal for real-time data ingestion pipelines, which align with the requirements for real-time analytics.
- C. Incorrect.
Incorrect: While Cloud Spanner is a managed database service, it is not designed for ad-hoc queries or analytics workloads. It is more suitable for transactional use cases.
- D. Correct.
Correct: Data Catalog enables metadata management, which is crucial for improving data discoverability and governance in the new architecture.
- E. Incorrect.
Incorrect: A Lift-and-Shift approach using Compute Engine VMs would not address scalability or reduce operational overhead. This solution would replicate the current limitations of the on-premises system.