Google Professional Data Engineer Question 125
Select 2Google Cloud PlatformYour company is building a data pipeline to handle both real-time and batch data processing for its e-commerce platform. The pipeline needs to store transactional data, such as customer orders, and large volumes of clickstream data from the website. The transactional data requires low-latency access for order processing, while the clickstream data needs to be stored cost-effectively for long-term analytics. Which storage systems should you choose?
- A
Cloud Spanner for transactional data and Cloud Storage for clickstream data
- B
Bigtable for transactional data and Cloud SQL for clickstream data
- C
Cloud SQL for transactional data and BigQuery for clickstream data
- D
Firestore for transactional data and Cloud Storage for clickstream data
- E
BigQuery for transactional data and Firestore for clickstream data
Show answer and explanation
Correct answers: A, C
Explanation
Selecting the appropriate storage system depends on the specific use case and data characteristics. For transactional data requiring strong consistency and low latency, Cloud Spanner or Cloud SQL are good choices. For clickstream data, which involves large-scale unstructured data for analytics, Cloud Storage or BigQuery is ideal. Cloud Spanner and Cloud Storage, or Cloud SQL and BigQuery, are the best combinations for this scenario.
- A. Correct.
Correct. Cloud Spanner is a globally distributed, horizontally scalable relational database ideal for transactional data requiring strong consistency and low latency. Cloud Storage is suitable for storing large volumes of unstructured data like clickstream logs cost-effectively.
- B. Incorrect.
Incorrect. Bigtable is optimized for low-latency, high-throughput use cases but is not ideal for handling transactional data with strong consistency requirements. Cloud SQL is also not suited for large-scale, long-term storage of clickstream data.
- C. Correct.
Correct. Cloud SQL is a managed relational database service that works well for transactional data requiring SQL capabilities. BigQuery is suitable for storing and analyzing large datasets like clickstream data.
- D. Incorrect.
Incorrect. Firestore is a NoSQL database optimized for hierarchical, document-based data but may not meet the low-latency transactional requirements of order processing. Cloud Storage is appropriate for clickstream data but does not fit the transactional use case.
- E. Incorrect.
Incorrect. BigQuery is designed for analytical workloads, not transactional processing. Firestore is not suited for storing and analyzing large volumes of clickstream data.