Cloud Digital Leader Google Cloud Products and Services Practice Test 2 — 15 Questions

Question 1

A company uses App Engine Standard environment for a Python web app. During a traffic spike, they notice increased latency. They want to improve performance without changing code. What should they do?

Accepted Answer

Increase the minimum idle instances in automatic scaling settings. To reduce latency during traffic spikes without changing code, increase the minimum number of idle instances (min_idle_instances) in automatic scaling settings. This keeps warm instances ready and avoids cold starts. Increasing maximum idle instances alone does not force instances to stay warm; it only raises the cap. Changing machine type is not available in App Engine Standard, migrating to Flex is a larger change, and Cloud CDN only caches static content.

Answer

Configure manual scaling with a larger machine type

Answer

Migrate to App Engine Flexible Environment

Answer

Enable Cloud CDN in front of the app

Question 2

A DevOps team wants to implement infrastructure as code for deploying a multi-service application on GKE. They need to store container images, automate builds on code commits, and manage Kubernetes YAML configurations. Which TWO services should they use? (Choose 2)

Accepted Answer

Cloud Build. Artifact Registry stores container images, Cloud Build automates builds on commits, and Deployment Manager is for managing cloud resources, not Kubernetes YAML configurations. For managing Kubernetes configs, tools like Config Sync or kubectl are used. Cloud Source Repositories is for source code, and Cloud Run is serverless. Thus, only Artifact Registry and Cloud Build directly fulfill the requirements.

Answer

Deployment Manager

Answer

Cloud Source Repositories

Answer

Cloud Run

Question 3

A startup wants to build a serverless event-driven application where a file upload to Cloud Storage triggers a function that sends a notification email. Which TWO services are essential? (Choose 2)

Accepted Answer

Cloud Functions. Cloud Functions is the only essential service: it can be triggered directly by Cloud Storage events to send the notification email. Cloud Pub/Sub is optional for decoupling or asynchronous processing but not required for the basic described functionality. Cloud Scheduler is for cron jobs, Cloud Tasks is for task queues, and Cloud Run is not needed for this event-driven trigger.

Answer

Cloud Scheduler

Answer

Cloud Tasks

Answer

Cloud Run

Answer

Cloud Pub/Sub

Question 4

A company needs to store petabytes of time-series IoT sensor data and query it with single-digit millisecond latency at millions of reads per second. The data has a simple key-value structure with timestamps. Which Google Cloud database is MOST appropriate?

Accepted Answer

Cloud Bigtable. Cloud Bigtable is designed for petabyte-scale, low-latency (single-digit ms), high-throughput NoSQL storage for time-series, IoT, and financial data. It scales horizontally by adding nodes. BigQuery is optimised for analytics (seconds-to-minutes latency), Cloud Spanner is a relational OLTP database with strong consistency but not optimized for petabyte-scale time-series key-value access at this latency/throughput, and Firestore is for document data with hierarchical structure.

Answer

BigQuery

Answer

Cloud Spanner

Answer

Firestore

Question 5

A company has a critical application running on Compute Engine instances in a managed instance group (MIG) that must remain highly available across three zones in a region. The application stores state on Persistent Disk. What is the correct approach to achieve high availability?

Accepted Answer

Use a regional Persistent Disk and attach it to instances in the regional MIG. For high availability across three zones, the storage must be accessible from every zone and continue operating if any one zone fails. A regional Persistent Disk spans exactly two zones and can be attached to only a single Compute Engine instance concurrently. Therefore, using a regional PD with a regional MIG does not satisfy the three-zone availability requirement. The correct approach would require a distributed storage solution (e.g., Cloud Spanner or multi-regional database) or application-level replication across the three zones, but none of the options provide that.

Answer

Use zonal Persistent Disk and snapshot to Cloud Storage every hour

Answer

Use Cloud Load Balancing without any disk changes

Answer

Use a regional managed instance group with a multi-regional bucket for state

Question 6

A company needs to replicate data between two Cloud SQL for PostgreSQL instances in different regions for disaster recovery. They require automated failover with minimal data loss. Which feature should they enable?

Accepted Answer

Cross-region read replica. Cloud SQL for PostgreSQL supports cross-region read replicas using asynchronous replication. These replicas can be promoted to standalone instances in a disaster scenario, providing a way to fail over with some data loss (replication lag). However, promotion is a manual operation, not automated. Point-in-time recovery relies on backups and does not provide real-time replication. Multi-region instances are not available for Cloud SQL; cross-region replicas are the correct approach for DR. Private IP addresses are for connectivity, not replication.

Answer

Point-in-time recovery with backups

Answer

Multi-region Cloud SQL instance

Answer

Private IP connectivity

Question 7

A company needs to store petabytes of time-series IoT sensor data and query it with single-digit millisecond latency at millions of reads per second. The data has a simple key-value structure with timestamps. Which Google Cloud database is MOST appropriate?

Accepted Answer

Cloud Bigtable. Cloud Bigtable is designed for exactly this use case — petabyte-scale, low-latency (single-digit ms), high-throughput NoSQL storage for time-series, IoT, and financial data. It scales horizontally by adding nodes. BigQuery is optimised for analytics (seconds-to-minutes latency), Cloud Spanner is a strongly consistent relational database for OLTP workloads, and Firestore is for document data for mobile and web apps.

Answer

Firestore

Answer

BigQuery

Answer

Cloud Spanner

Question 8

A company wants to build a global application that requires strong consistency across regions and an SLA of 99.999% for its transactional database. The workload involves moderate write throughput (10K writes/sec) and high read throughput (100K reads/sec). Which database should they choose?

Accepted Answer

Cloud Spanner. Cloud Spanner is a globally distributed relational database that provides ACID transactions across regions, strong consistency, and a 99.999% availability SLA. It can handle millions of reads per second with automatic horizontal scaling. Bigtable does not offer SQL or ACID transactions; Cloud SQL is not globally distributed; Firestore, while strongly consistent, is a NoSQL database and its multi-region locations are not truly global like Spanner's, and it lacks the relational transaction capabilities that Spanner offers for this use case.

Answer

Cloud SQL

Answer

Firestore

Answer

Cloud Bigtable

Question 9

A company wants to migrate an on-premises PostgreSQL database to Google Cloud with minimal changes to the application code. They also require high availability with automatic failover. Which service should they use?

Accepted Answer

Cloud SQL. Cloud SQL for PostgreSQL is a fully managed service that is wire-compatible with PostgreSQL, so it requires minimal application changes. It provides high availability with automatic failover and is the simplest choice for migrating an on-premises PostgreSQL database. AlloyDB is also PostgreSQL-compatible and provides HA, but it is designed for more demanding workloads and is not the default choice for a minimal-change migration. Cloud Spanner is not fully PostgreSQL-compatible, and Compute Engine requires manual HA setup.

Answer

AlloyDB

Answer

Cloud Spanner

Answer

Compute Engine with PostgreSQL

Question 10

A company needs to store files that are accessed infrequently (once a quarter) and wants the lowest storage cost. Data retrieval can take a few hours. Which Cloud Storage class should they use?

Accepted Answer

Archive. Archive storage is the lowest-cost storage class for data accessed less than once a year, with retrieval times up to hours. While Coldline is designed for data accessed less than once a quarter, Archive offers even lower cost and still meets the retrieval time requirement of a few hours. Nearline is for data accessed less than once a month, and Standard is for frequently accessed data.

Answer

Nearline

Answer

Coldline

Answer

Standard

Question 11

A company is running a stateful workload on Compute Engine and needs persistent block storage that can be attached to multiple VMs in read-only mode. Which two Google Cloud storage solutions meet this requirement?

Accepted Answer

Persistent Disk (with read-only replica). Persistent Disk (PD) is a block storage solution that can be attached to multiple VMs in read-only mode. Both standard Persistent Disk and Persistent Disk with a read-only replica support this capability, making them suitable for sharing stateful data across multiple instances in the same zone.

Answer

Filestore

Answer

Local SSD

Answer

Cloud Storage

Question 12

A startup is building a mobile app backend that stores user profiles and session data. They need a NoSQL database with real-time synchronization across devices and offline support. Which TWO Google Cloud services are suitable?

Accepted Answer

Firestore. Firestore is a NoSQL document database that provides real-time listeners and offline persistence, making it ideal for mobile app backends requiring synchronization across devices and offline support. Cloud Storage is an object storage service, not a NoSQL database, and does not offer real-time sync or offline database capabilities. None of the other options (Memorystore, Cloud Bigtable, Cloud Spanner) meet the specific requirements of a NoSQL database with real-time sync and offline support.

Answer

Memorystore

Answer

Cloud Bigtable

Answer

Cloud Spanner

Answer

Cloud Storage

Question 13

A startup wants to build a chatbot that can understand natural language and integrate with Google Workspace (e.g., Gmail, Calendar). They have limited ML expertise. Which TWO Google Cloud services should they use?

Accepted Answer

Dialogflow. Dialogflow is the managed Google Cloud service for building conversational interfaces. It provides natural language understanding, conversation management, and Google Workspace integrations while requiring minimal ML expertise. Cloud Functions is the serverless compute service used for webhook fulfillment, allowing the chatbot to call Google Workspace APIs such as Gmail and Calendar. Vertex AI is a general-purpose ML platform, Cloud Translation API translates text, and Natural Language API provides text analysis, but none of these directly builds and integrates a chatbot.

Answer

Vertex AI

Answer

Cloud Translation API

Answer

Natural Language API

Question 14

A company needs to store petabytes of time-series IoT sensor data and query it with single-digit millisecond latency at millions of reads per second. The data has a simple key-value structure with timestamps. Which Google Cloud database is MOST appropriate?

Accepted Answer

Cloud Bigtable. Cloud Bigtable is designed for petabyte-scale, low-latency, high-throughput NoSQL storage for time-series, IoT, and financial data. It scales horizontally by adding nodes.

Answer

Cloud Spanner

Answer

BigQuery

Answer

Firestore

Question 15

A developer wants to deploy a containerized web application that can automatically scale to zero when there are no requests, and charges only for resources used during request processing. Which Google Cloud compute service should they use?

Accepted Answer

Cloud Run. Cloud Run is a serverless container platform that scales to zero and charges per request. Google Kubernetes Engine and Compute Engine require running instances, and App Engine Standard is a platform as a service but not container-based.

Answer

Compute Engine

Answer

Google Kubernetes Engine (GKE)

Answer

App Engine Standard Environment