Skip to content

Latest commit

 

History

History
executable file
·
406 lines (292 loc) · 18.2 KB

File metadata and controls

executable file
·
406 lines (292 loc) · 18.2 KB

Google Cloud Platform (GCP)

Q1. You want to schedule a cron job to send a message to a Cloud Pub/Sub topic hourly. What do you do?

  • Create a Cloud Function with a target of HTTP
  • Create a Cloud Scheduler job with a target of HTTP
  • Create a Cloud Function with a target of Cloud Pub/Sub
  • Create a Cloud Scheduler job with a target of Cloud Pub/Sub

Q2. You want to set up a managed Apache Spark cluster. Which service do you configure?

  • Cloud Dataprep
  • Cloud Composer
  • Cloud Dataflow
  • Cloud Dataproc

Q3. GCP Organizations are not available in your GCP project. What type of account are you using?

  • Temporary
  • Free Tier
  • Local
  • Organizational

Note: Free trial doesn't have the organizations.

Q4. What feature do you need to enable to support load balancing and autoscaling managed instance groups for Compute Engine?

  • metadata
  • security scans
  • network endpoint groups
  • health checks

Q5. You are designing an application that will send messages with the MQTT protocol. What service should you use?

  • BigQuery
  • Cloud Pub/Sub
  • IoT Broker
  • Bigtable

Q6. You want to configure services to host a mobile application that uses TensorFlow Lite models. Which services should you use?

  • Cloud Functions and ML Kit
  • Firebase and AI Platform jobs
  • Firebase and ML Kit
  • Cloud Functions and AI Platform jobs

Q7. You need to deploy Redis with unique persistent identities and stable hostnames on GKE. What type of deployment object do you use?

  • GCP deployment
  • YAML configuration file
  • Python nested template
  • stateful set

Q8. You want to try out a machine-learning TensorFlow example notebook with the fewest configuration steps on GCP. What do you do?

  • Upload the notebook to Google Colaboratory. Set the hardware accelerator to use GPU.
  • Create a notebook instance with TPU. Upload the notebook to the instance.
  • Create a notebook instance with GPU. Upload the notebook to the instance.
  • Upload the notebook to Google Colaboratory. Set the hardware accelerator to use TPU.

Q9. You need to configure a data retention policy for a Cloud Storage bucket to govern how long objects in the bucket must be retained. What feature do you enable?

  • Requester Pays
  • Bucket Lock
  • Bucket Policy Only
  • Encryption Type

Q10. To be able to configure an existing Cloud Function, what information should you review and from which location?

  • execution time in Cloud Storage logs
  • execution time in Stackdriver logs
  • text payload in Stackdriver logs
  • text payload in Cloud Storage logs

Q11. You need to execute a "lift and shift" move with as few steps as possible. Which items do you need to create to accomplish this task, listed in the correct order of creation?

  • 1. Create a VPN from GCP to your corporate network. 2. Create a GCP network for the project. 3. Create the VM instances to support the application.
  • 1. Create a GCP network for the project. 2. Create a VPN from your corporate network to GCP. 3. Create the VM instances to support the application.
  • 1. Create a GCP network for the project. 2. Create a VPN from GCP to your corporate network. 3. Create the VM instances to support the application.
  • 1. Create a VPN from your corporate network to GCP. 2. Create a GCP network for the project. 3. Create the VM instances to support the application.

Q12. You are concerned about data egress charges for files in a Cloud Storage bucket. What configuration change should you make?

  • Enable Requester Pays on the project.
  • Enable Requester Pays on the bucket.
  • Change the default storage class from multi-regional to regional.
  • Change the default storage class from regional to multi-regional.

Q13. You need to troubleshoot an Android app that uses Cloud Storage. Where do you look for logs?

  • Firebase Crashlytics
  • Firebase Crashlytics and Stackdriver
  • Firebase Testlab and Stackdriver
  • Stackdriver

Q14. You want to configure services to be able to perform a regular interactive query of Google AdWords data. What should you do?

  • Plan to use BigQuery. Create a BigQuery data set to hold the data.
  • Plan to use Cloud SQL. Enable Cloud SQL Transfer Data API.
  • Plan to use Cloud SQL. Create a Cloud SQL Database to hold the data.
  • Plan to user BigQuery. Enable BigQuery Transfer Data API.

Q15. You want to use GCP-preferred base container images. Where can your users find these images?

  • Docker Hub
  • Compute Engine base images
  • GCP Marketplace
  • Google Kubernetes Engine base images

Q16. You want to set a notification for service spend in GCP. What do you do, and who will be notified?

  • Use the GCP Billing console Budgets section to set a billing alert. The billing admin will be notified if the alert threshold is exceeded.
  • Use the GCP Billing console Budgets section to set a billing alert. The organizational admin will be notified if the alert threshold is exceeded.
  • Use the GCP Organizational console Budgets section to set a billing alert. The project admin will be notified if the alert threshold is exceeded.
  • Use the GCP Organizational console Budgets section to set a billing alert. The organizational admin will be notified if the alert threshold is exceeded.

Q17. Cloud SQL supports which two relational databases?

  • Redis and Oracle
  • MySQL and Oracle
  • Redis and PostgreSQL
  • MySQL and PostgreSQL

Q19. You want to quickly test a single Docker container image that will start a stateless HTTP service on GCP. Which service do you select to host your container image?

  • Cloud Run
  • Cloud Run on Google Kubernetes Engine
  • Cloud Functions
  • Google Kubernetes Engine

Q20. You want to develop Firebase security rules following Google best practices. Which tool(s) do you use?

  • Firebase Emulator
  • Firebase Emulator and GCP Rules Simulator
  • GCP policy emulator and Firebase Rules Simulator
  • Firebase Emulator and Firebase Rules Simulator

Q21. You want to build an application with conversation interpretation based on machine learning speech understanding. Which service is a best fit for this scenario?

  • Dialogflow
  • Natural Language Translation
  • Speech API
  • AutoML Natural Language

Q22. Which service must you enable on your GKE cluster in order to use Cloud Run on Google Kubernetes Engine (GKE)?

  • Stackdriver
  • Istio
  • Knative
  • Spinnaker

Q23. How can you enable clients inside a GKE cluster to be able to contact Pods?

  • Configure the clients to connect via the default cluster IP.
  • Configure an external name. Configure the clients to connect via the external name.
  • Configure a node port. COnfigure the clients to connect via the node port.
  • Configure a load balancer. Configure the clients to connect via the load balancer.

Q24. You want to capture the message acknowledgment state of a subscription at a given time. What do you do?

  • Create a Cloud Pub/Sub subscription backup.
  • Create a Cloud Pub/Sub subscription topic.
  • Create a Cloud Pub/Sub subscription snapshot.
  • Create a Cloud Pub/Sub subscription.

Q25. You need to set a bucket storage class for data that will be accessed once a month or less. Which type do you select?

  • coldline
  • multi-regional
  • regional
  • nearline

Q26. Which command copies a file named test.txt to a Cloud Storage bucket named myBucket?

  • gcloud cp gs://myBucket test.txt
  • gsutil cp gs://myBucket test.txt
  • gcloud cp test.txt gs://myBucket
  • gsutil cp test.txt gs://myBucket

Q27. Which command adds a column family named cf1 to your Bigtable table named my-table?

  • cbt createfamily cf1 my-table
  • cbt createfamily my-table cf1
  • gcloud createfamily cf1 my-table cf1
  • gcloud createfamily my-table cf1

Q28. Which command do you use to connect to a running Compute Engine instance with SSH?

  • gcloud ssh compute [instance_name] --zone=[zone]
  • compute ssh connect [instance_name] --zone=[zone]
  • gcloud compute ssh [instance_name] --zone=[zone]
  • compute connect ssh [instance_name] --zone=[zone]

Q29. You need to set CORS for some objects in a bucket for a client-side application. What do you do?

  • Segment the objects into two buckets. For the CORS objects, set the CORS policy at the bucket level.
  • Segment the objects into two buckets. For the CORS objects, set the CORS policy at the bucket level. Add client-side code to handle the requests.
  • Leave the objects in one bucket. Set the CORS policy at the object level.
  • Leave the objects in one bucket. Set the CORS policy at the object level. Add client-side code to handle the requests.

Q30. For your application, you are building a CI/CD pipeline that logs information when a message is sent to a queue. Which services should you select?

  • Cloud Functions and Cloud Spanner
  • Cloud Functions and Cloud Pub/Sub
  • Cloud Run and Cloud Pub/Sub
  • Cloud Run and Cloud Spanner

Q31. You need to create instances with identical configurations that cannot be changed. Which Compute Engine feature do yo use?

  • instance templates
  • instance groups
  • snapshots
  • instances

Q32. You want to automatically perform security scans on user-developed containers. Where should you advise your users to host their containers?

  • Google Container Registry
  • GitLab
  • Artifactory
  • Docker Hub

Q33. You need to control incoming and outgoing traffic to a Compute Engine instance. What do yo do?

  • Configure firewall rules.
  • Configure VPC Network Peering.
  • Configure routes.
  • Configure external IP addresses.

Q34. Which object is a container for GCP services instances?

  • group
  • folder
  • project
  • organization

Q35. What is needed to access GCP APIs in your project?

  • GCP VMs
  • GCP buckets
  • GCP credentials
  • GCP users

reference link

Q36. Which object is used to hold payment information for a GCP project?

  • organization
  • billing account
  • folder
  • payment account

reference link

Q37. You need to cache your HTTP and HTTPS static content closer to users, so that you can get faster delivery and reduce serving costs. Which service do you set up?

  • Cloud DNS
  • Cloud NAT
  • Cloud CDN
  • Cloud Load Balancing

reference link

Q38. You are designing security for an IoT device application. What best practices should you include in your design?

  • Use a private key for each device. Keep the private key secret.
  • Use a public key for each device. Keep the public key secret.
  • Use a public private key pair for each device. Keep the public key secret.
  • Use a public private key pair for each device. Keep the private key secret.

reference link

Q39. You need to set a WordPress site with high availability on GCP using the fewest steps. What do you do?

  • Configure and launch the WordPress HA deployment template from the GCP Marketplace.
  • Configure and launch the WordPress HA deployment template from the WordPress Marketplace.
  • Create, configure, and launch a WordPress HA deployment template on Compute Engine.
  • Create, configure, and launch a WordPress HA deployment template on Google Kubernetes Engine.

Q41. How do you configure Cloud Build to rebuild your image when a change is made to the source code?

  • Add a Cloud Build trigger, and set it to fire on commit to associate repository.
  • Add a Cloud Build function, and set it to fire on commit to associate repository.
  • Add a Cloud Build trigger, and set it to fire on commit to associate registry.
  • Add a Cloud Build function, and set it to fire on commit to associate registry.

Q42. To be able to decrypt files that were encrypted with a customer-manager key, what conditions must be met?

  • Specify the decryption key with the same name as the encryption key. At least one version of a key with that name must be enabled.
  • Specify a decryption key and version with the same name as the encryption key and version. The name key and that version must be enabled.
  • Specify the decryption key with the same name as the encryption key. The named key and that version must be enabled.
  • Specify a decryption key and version with the same name as the encryption key and version. At least one version of a key with that name must be enabled.

Q43. You want to encrypt a Compute Engine. What do you need to do?

  • Nothing. Images are encrypted by default with a Google-managed key.
  • Nothing. Images are encrypted by default with a customer-managed key.
  • Associate you image with a customer-managed key to encrypt it.
  • Associate you image with a Google-managed key to encrypt it.

Reference

Q44. You want to quickly evaluate Cloud Filestore. What are the steps, in order, to set up a test environment?

  • 1. Create a Filestore instance.
    2. Mount the fileshare from that instance on a client VM instance.
    3. Create a Compute Instance VM to be the client.
    4. Create a file on the mounted fileshare.
    
  • 1. Create a Filestore instance.
    2. Mount the fileshare from that instance on a client VM instance.
    3. Create a file on the mounted file share.
    4. Create a Compute Instance VM to be the client.
    
  • 1. Create a Compute Instance VM to be the client.
    2. Create a Filestore instance.
    3. Mount the file share from that instance on a client VM instance.
    4. Create a file on the mounted fileshare.
    
  • 1. Create a Compute Instance VM to be the client.
    2. Create a Cloud Filestore instance.
    3. Mount the fileshare from that instance on a client VM instance.
    4. Create a file on the mounted file share.
    

Reference

Q45. You need to optimize spending for always-on Compute Engine virtual machines of certain sizes and types-with minimum effort. What do you do?

  • Analyze machine usage to verify that machines will be on for more that 25% of the time, and verify that contained use discounts have been automatically applied to your bill.
  • Analyze machine usage to verify that machines will be on for more that 25% of the time, and verify that sustained use discounts have been automatically applied to your bill.
  • Analyze machine usage to verify that machines will be on for more than 25% of the time and purchase contained use discounts.
  • Analyze machine usage to verify that machines will be on for more than 25% of the time and purchase sustained use discounts.

Reference

Q46. You are designing a GCP replacement for your on-premises data warehouse. Which service do you select?

  • Cloud SQL
  • Bigtable
  • Cloud Spanner
  • BigQuery

Q47. How can you quickly back up a persistent disk on Compute Engine?

  • Take a snapshot
  • Create an instance group
  • Create an instance template
  • Create an image

Q48. You are implementing IAM roles for your organization following GCP best practice. Which statement best reflects this?

  • At minimum, you should have two levels of access: one for admins, one for users.
  • At minimum, you should have three levels of access: one for networking, one for security, one for developers.
  • At minimum, you should have three levels of access: one for admins, one for security, one for users.
  • At minimum, you should have two levels of access: one for networking, one for developers.

Q49. You need to quickly build an application that logs all file uploads to a Cloud Storage bucket. Which service do you select?

  • Cloud Functions
  • Cloud Run
  • Google Kubernetes Engine
  • Google Compute Engine

Reference

Q50. You need to answer the question "How long does it take my application to handle a given request?" What do you do?

  • Configure Stackdriver Error Reporting.
  • Configure Stackdriver Profiler.
  • Configure Stackdriver Trace.
  • Configure Stackdriver Debug.

Reference

Q51. You want to select a fully managed, serverless platform for developing and hosting web applications at scale on GCP. Which service should you choose?

  • Google App Engine.
  • Cloud Functions.
  • Google Kubernetes Engine.
  • Google Compute Engine.

Q52. You have a set of exploratory queries that you want to run as inexpensively as possible in BigQuery. What should you do?

  • Set the job priority for each query to Batch and expect to see query results immediately.
  • Set the job priority for each query to Interactive and expect to see query results immediately.
  • Set the job priority for each query to Batch and expect to see query results within 24 hours.
  • Set the job priority for each query to Batch and expect to see query results within 24 hours.

Explanation: BigQuery offers a choice of two pricing models for runnig queries: on-demand (you are charged for the number of bytes processes by each query) and flat-rate (you purchase slots - dedicated processing capacity). If the exploratory queries are one-off, running in Interactive mode using on-demand pricing seems to be the optimal answer.

Reference

Q53. How do you run a Docker container image on a Google Compute Engine virtual machine?

  • Create a VM instance, and then SSH into the instance and install Kubernetes tools.
  • Create a VM instance, and then SSH into the instance and install Docker tools.
  • Create a VM instance and, in the web UI, select Allow HTTP traffic.
  • Create a VM instance and, in the web Ul, select Deploy a container image to this instance.