Answer-first summary for fast verification
Answer: Replace the existing data warehouse with BigQuery. Use table partitioning.
The question requires designing a reliable and scalable data warehouse solution for TerramEarth on GCP. Based on the community discussion and best practices, option A (Replace the existing data warehouse with BigQuery using table partitioning) is the optimal choice. BigQuery is a serverless, highly scalable data warehouse that aligns with TerramEarth's needs for handling large volumes of time-series data. Table partitioning improves query performance and cost efficiency by organizing data into manageable segments (e.g., by date), which is crucial for time-series analytics. The community consensus strongly supports A, with 88% of votes and high upvotes for comments emphasizing its reliability and performance benefits. Option C (using federated data sources) is less suitable due to performance limitations, higher costs for external data access, and constraints like concurrent query limits, making it unreliable for a core data warehouse. Options B and D, which involve Compute Engine instances, are not scalable or cost-effective compared to BigQuery's serverless architecture. Thus, A is the best approach for meeting TerramEarth's requirements.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
As a certified Google Professional Cloud Architect working on the TerramEarth case study, you are tasked with designing a reliable and scalable data warehouse solution on GCP. Based on TerramEarth's business and technical requirements, what is your recommended approach?
A
Replace the existing data warehouse with BigQuery. Use table partitioning.
B
Replace the existing data warehouse with a Compute Engine instance with 96 CPUs.
C
Replace the existing data warehouse with BigQuery. Use federated data sources.
D
Replace the existing data warehouse with a Compute Engine instance with 96 CPUs. Add an additional Compute Engine preemptible instance with 32 CPUs.
No comments yet.