Google Professional Data Engineer Question 167
Select 3Google Cloud PlatformYour organization is building a data mesh architecture to enable decentralized data ownership while maintaining a centralized governance model. You are tasked with implementing a solution using Google Cloud tools. The requirements include:
- Ensure data discoverability and metadata management across all datasets.
- Provide a way for domain teams to securely store and manage their data.
- Enable domain teams to query data from other domains without requiring data duplication.
Which combination of tools would best meet these requirements?
- A
Use Dataplex for data governance and discovery, and BigQuery for querying datasets across domains.
- B
Use Cloud Storage for storing data by domain teams and Data Catalog for metadata management.
- C
Use BigQuery with authorized views to enable secure cross-domain data access without duplication.
- D
Use Dataproc for metadata management and Cloud SQL for storing datasets.
- E
Use Dataplex for centralized governance and Dataflow for data discoverability.
Show answer and explanation
Correct answers: A, B, C
Explanation
A data mesh architecture requires tools that support decentralized data ownership while maintaining centralized governance and discoverability. Dataplex enables governance and discovery, Data Catalog ensures metadata management, Cloud Storage provides secure domain-specific data storage, and BigQuery facilitates querying across domains securely. Using these tools together satisfies the requirements of the scenario.
- A. Correct.
Correct: Dataplex provides centralized governance and data discovery, while BigQuery allows querying datasets across domains without duplication, which is essential in a data mesh.
- B. Correct.
Correct: Cloud Storage enables domain teams to securely store their data, and Data Catalog provides metadata management and discoverability across datasets.
- C. Correct.
Correct: BigQuery authorized views enable secure access to data from other domains without duplicating the data, aligning with the principles of a data mesh.
- D. Incorrect.
Incorrect: Dataproc is primarily used for big data processing, not metadata management, and Cloud SQL is not well-suited for large-scale domain-specific data storage.
- E. Incorrect.
Incorrect: While Dataplex is useful for governance, Dataflow is primarily a data processing tool and does not provide data discoverability features.