In a retrieval-augmented generation (RAG) pipeline, what is the primary role of a vector database?

To train the large language model

To store the large language model's parameters

To store embeddings of documents and retrieve relevant ones for the LLM

To generate the embeddings for the documents

An IoT company is collecting sensor data from millions of devices. The data is time-series and write-heavy, with occasional reads for monitoring. The team needs a database that can handle high write throughput and scale horizontally. Which database solution is the best fit?

Use Cloud SQL for PostgreSQL with read replicas

Use Bigtable, which is a managed NoSQL database optimized for high write throughput

Use BigQuery for real-time data ingestion

Use Firestore, which is a managed NoSQL document database with automatic scaling

A social media company is building a feature that recommends friends based on mutual connections. The data is highly connected and the queries involve traversing multiple levels of relationships. The team also needs to store user profiles with flexible attributes. Which database solution is the best fit?

Use a graph database like Neo4j on Compute Engine for relationships and Firestore for profiles

Use Cloud SQL for PostgreSQL with recursive queries for graph traversal

Use Firestore to store user profiles and perform graph traversal in the application

Use BigQuery to store the graph and run SQL queries for traversal

A gaming company is launching a new multiplayer game with a highly variable and unpredictable workload. They need a database that can handle millions of concurrent users and scale horizontally without downtime. The company has a policy that requires all data to be encrypted at rest and in transit, and they want to minimize operational overhead. Which database solution is the best choice?

Use Bigtable with automatic scaling and encryption enabled

Use Cloud Spanner with automatic scaling and encryption enabled

Use Cloud SQL for PostgreSQL with read replicas and manual scaling

Use Firestore with automatic scaling and encryption enabled

Google CloudProfessional Cloud Database Engineer

Domain 1Objective 4

1.4 Evaluate Appropriate Database Solutions on Google Cloud PROFESSIONAL-CLOUD-DATABASE-ENGINEER Practice Questions (Page 1)

Part of the Design innovative, scalable, and highly available cloud database solutions domain, which accounts for ~32% of the PROFESSIONAL-CLOUD-DATABASE-ENGINEER exam.

31questions here

7free pages

8concepts

~32%of the exam

8 Concepts in This Objective

Managed vs. Unmanaged Database Services
SQL vs. NoSQL Business Requirements
Cost Analysis of Database Solutions
Application and Database Dependencies
Regulatory and Compliance Requirements
Organizational Policies Impact on Database Strategy
Multi-Database Technology Solutions
Database Support for Generative AI and LLMs

Questions 1–5

1foundation · easy
In a retrieval-augmented generation (RAG) pipeline, what is the primary role of a vector database?
Select an answer first
2application · medium
An IoT company is collecting sensor data from millions of devices. The data is time-series and write-heavy, with occasional reads for monitoring. The team needs a database that can handle high write throughput and scale horizontally. Which database solution is the best fit?
Select an answer first
3foundation · easy
An application needs to query data that is stored in both a relational database and a NoSQL database. Which approach allows the application to query both as if they were a single source?
Select an answer first
4expert · hard
A social media company is building a feature that recommends friends based on mutual connections. The data is highly connected and the queries involve traversing multiple levels of relationships. The team also needs to store user profiles with flexible attributes. Which database solution is the best fit?
Select an answer first
5expert · hard
A gaming company is launching a new multiplayer game with a highly variable and unpredictable workload. They need a database that can handle millions of concurrent users and scale horizontally without downtime. The company has a policy that requires all data to be encrypted at rest and in transit, and they want to minimize operational overhead. Which database solution is the best choice?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to PROFESSIONAL-CLOUD-DATABASE-ENGINEER

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Google Cloud. “PROFESSIONAL-CLOUD-DATABASE-ENGINEER” is a trademark of its owner, used for identification only.

1.4 Evaluate Appropriate Database Solutions on Google Cloud PROFESSIONAL-CLOUD-DATABASE-ENGINEER Practice Questions (Page 1)

Questions 1–5

In a retrieval-augmented generation (RAG) pipeline, what is the primary role of a vector database?

An IoT company is collecting sensor data from millions of devices. The data is time-series and write-heavy, with occasional reads for monitoring. The team needs a database that can handle high write throughput and scale horizontally. Which database solution is the best fit?

An application needs to query data that is stored in both a relational database and a NoSQL database. Which approach allows the application to query both as if they were a single source?

Keep studying — Design innovative, scalable, and highly available cloud database solutions