Google Cloud Associate Practice Exam Questions and Answers
Q1. How you can use service account say “testonesa” which will enable access to required other services from Compute Engine.
- gcloud auth activate-service-account [service account] –key-file [keyfilenae
Q2. BigQuery can get expensive if it processing huge data, the best practice is to estimate each query before you execute it. Which of the following command will help you to get an estimation.
- bq –location=[LOCATION] query –dry_run ‘SQL QUERY’ Limit 10
Q3. You are tasked to identify the best solution for the analytics databases which is low latency, high throughput for visualizations. which of the following is your best choice?
- Cloud BigTable
Q4.You are tasked to perform maintenance of Cloud Bigtable in production. Nowadays the performance degraded and you will need to work on how to improve the performance. While the development team is working on query performance, what will be your quick solution to improve performance?
- Add Nodes as the performance of Cloud Bigtable increases with nodes
Q5. Social community application is deployed in App Engine, customers are complaining slow performance of some pages. You started looking into issues – Which of the following application will help you to understand the latency of different endpoints.
- Stackdriver Trace
Q6. Its best practice to use cache data for queries runs on BigQuery frequently. but you can disable cached data for some queries.
Q7. You can improve the cost of Cloud BigQuery using all of the following methods except one.
- Use min number of nodes while processing the data
Q8. The company sells graphics design using its website, Currently, there is a process to deliver design in email, and many times customers lose emails and they call to care. The call to the care has been increased because many times emails are going into spam.
The business did some analysis and came up with an idea customers can get links to design as well as trigger email using the website. Business does not want the user to have access to Cloud Storage permanently it should be configurable for a short duration (e.g. 24 hours – 7 days ) so that they can download. What solution do you suggest to the developer to use it?
- Create a Signed URL for objects with a specific expiry date/time on them and share it with the user.
Q9. You need to create the project in google cloud using CLI’s inside organization 12345, Which of the following command will be useful.
- $ gcloud projects create example-3 –organization=12345
Q10. You will need to upload the “data.txt” -> 30GB file to cloud storage, What will you do to make sure the transfer is resumed able and can be done in parallel if required.
- gsutils will make sure the transfer is resumable without any additional configuration. To Transfer it in parallel – it uses -m. gsutil -m cp -p file gs://bucket/obj
Q11. You are uploading data from the local server to Google Cloud Storage. You enabled parallelism with the m-switch. if the performance is not improved what you think it should be. You want to increase no of parallel uploads and see the performance. What options do you have?
- You can configure the file and set environment BOTO_PATH for file location or supply the following parameters at CLI uploads.gsutil o “GSUtil:parallel_process_count=8” -o “GSUtil:parallel_thread_count=1” \
Q12.You have application logs getting uploaded daily to Google Cloud Storage. You are tasked to convert the batch uploads to stream it to Cloud Storage. Which of the following is the best option you will choose?
- Cloud Storage does support streaming uploads.
Q13.You are tasked to manage the data archival process. You want to make sure that data is copied to Storage where it is used daily for 6 months and then used once in a month and after a year used once a year and after five years it may or may not be used. After 10 years it’s not required anymore. What will be your solution satisfying the above requirement keeping costs low?
- Use Cloud Storage to store data and use lifecycle policy to move data from different storage classes.
Q14.You have marketing websites that contain static HTML pages with images and videos. There is no dynamic content or programming to process data. All it has is java scripts that pull images or videos based on the browser. Currently, this website is hosted in the Datacenter server but it’s pretty slow. What is the best possible solution that you can propose keeping costs low with good performance.?
- Host complete website on Cloud Storage including images and videos.
Q15.The connected car company has all cars connected to the cloud and supply data continuous data in form of JSON messages. Backend cloud is implemented using Cloud Function and data is uploaded to BigQuery for Analytics. As the load has increased the cloud bill started getting increased and realized that Cloud Functions turn out to be expensive. Which of the following right choice to take data from all cars to big query.
- Cloud Pub/Sub
Q16.The company has cloud infrastructure in the google cloud platform. Launched and managed using deployment manager. Now they want to create the same infrastructure in other organizations which they acquired recently. How they can quickly launch Cloud Resources?
- Deployment Manager works based on templates. You can take the same template and use the deployment manager to launch cloud resources.
Q17.The company has multiple projects and many resources under each project. Many departments use different billing accounts to pay bills. Accounting has problems in consolidating and tracking different billing account data and suggested using a single billing account still report the use of cloud resources by different departments. What is your solution?
- Create a Single Billing Account and ask all departments to use the same billing account for all their projects. Take all cloud billing transactions for all projects to Cloud BigQuery to generate reports based on projects and departments.
Q18. The company has multiple departments and does specific functions. Cloud Resources permissions are becoming very complex to manage because people leave the organization, New people join or existing people change departments. What is the best way to manage permissions to cloud users in this case?
- Use Cloud Identify groups for different roles and assign specific permissions to groups via roles.
Q19.Which of the following service configuration Cloud CDN can be enabled?
- Load Balancer
Q20.A firewall is a default open for Google Compute Engine for SSH connection from the third-party client.
Q21.What is the purpose of a Cloud Router and why does that matter?
- It connects VPN on one side to Direct Peering on the other, which is faster than VPN alone.
Q22.What are the three types of networks offered in the Google Cloud Platform?
- Default network, auto network, and custom network.
Q23.A company requires a 500 Mbps connection with a secure channel to connect with GCP, Which one of the following interconnections is appropriate to keep cost low?
- Cloud VPN
Q24.True or False: NAT Instance used to connect VM from one network to connect VM in other networks which do not have external IP addresses
Q25.What is one benefit of applying firewall rules by tag rather than by address?
- When a VM is created with a matching tag, the firewall rules apply irrespective of the IP address it is assigned.
Q26.The HTTPS load balancer uses the following technique to send the same client traffic to the same backend instance.
- Session Affinity
Q27.All statements about Kubernetes Node Pool’s are true except one
- Node Pool is global
Q28.Instance templates are used to create a group of identical VMs. The instance templates include:
- A. Machine type, boot disk image or container image, zone, and labels
- B. Cloud Storage bucket definitions
- C. A load balancer description
- D. App Engine configuration file
Q29.The command-line command to create a Cloud Storage bucket is:
- A. gcloud MB
- B. gsutil MB
- C. gcloud mkbucket
- D. gsutil mkbucket
Q30. Your company has an object management policy that requires that objects stored in Cloud Storage is migrated from regional storage to nearline storage 90 days after the object is created. The most efficient way to do this is to:
- A. Create a cloud function to copy objects from regional storage to nearline storage.
- B. Set the MigrateObjectAfter property on the stored object to 90 days.
- C. Copy the object to persistent storage attached to a VM and then copy the object to a bucket created on nearline storage.
- D. Create a lifecycle management configuration policy specifying an age of 90 days and SetStorageClass as nearline.
Q31.An education client maintains a site where users can upload videos, and your client needs to assure redundancy for the files; therefore, you have created two buckets for Cloud Storage.
Which command do you use to synchronize the contents of the two buckets?
- A. gsutil rsync
- B. gcloud cp sync
- C. gcloud rsync
- D. gsutil cp sync
Q32.VPCs are resources.
- A. Regional
- B. Zonal
- C. Global
- D. Subnet
Q33.A remote component in your network has failed, which results in a transient network error.
When you submit a gsutil command, it fails because of a transient error. By default, the
- A. Terminate and log a message to Stackdriver
- B. Retry using a truncated binary exponential back-off strategy
- C. Prompt the user to decide to retry or quit
- D. Terminate and log a message to Cloud Shell
Q34.All of the following are components of firewall rules except which one?
- A. Direction of traffic
- B. Action on match
- C. Time to live (TTL)
- D. Protocol
Q35.Adding virtual machines to an instance group can be triggered in an autoscaling policy by
all of the following, except which one?
- A. CPU utilization
- B. Stackdriver metrics
- C. IAM policy violation
- D. Load balancing serving capacity
Q36.Your company’s finance department is developing a new account management application
that requires transactions and the ability to perform relational database operations using
fully compliant SQL. Datastore options in GCP include:
- A. Spanner and Cloud SQL
- B. Datastore and Bigtable
- C. Spanner and Cloud Storage
- D. Datastore and Cloud SQL
Q37.The marketing department in your company wants to deploy a web application but does
not want to have to manage servers or clusters. A good option for them is:
- A. Compute Engine
- B. Kubernetes Engine
- C. App Engine
- D. Cloud Functions
Q38.Your company is building an enterprise data warehouse and wants SQL query capabilities
over petabytes of data, but does not want to manage servers or clusters. A good option for
- A. Cloud Storage
- B. BigQuery
- C. Bigtable
- D. Datastore
Q39.You have been hired as a consultant to a startup in the Internet of Things (IoT) space. The
the startup will stream large volumes of data into GCP. The data needs to be filtered, transformed, and analyzed before being stored in GCP Datastore. A good option for the stream
the processing component is:
- A. Dataproc
- B. Cloud Dataflow
- C. Cloud Endpoints
- D. Cloud Interconnect
Q40.Preemptible virtual machines may be shut down at any time but will always be shut down
- A. 6 hours
- B. 12 hours
- C. 24 hours
- D. 48 hours
Q41.You have been tasked with designing an organizational hierarchy for managing departments and their cloud resources. What organizing components are available in GCP?
- A. Organization, folders, projects
- B. Buckets, directories, subdirectories
- C. Organizations, buckets, projects
- D. Folders, buckets, projects
Q42.During an incident that has caused an application to fail, you suspect some resource may
not have appropriate roles granted. The command to list roles granted to a resource is:
- A. gutil iam list-grantable-roles
- B. gcloud IAM list-grantable-roles
- C. gcloud list-grantable-roles
- D. gcloud resources grantable-roles
Q43.The availability of CPU platforms can vary between zones. To get a list of all CPU types
available in a particular zone, you should use:
- A. gcloud compute zones describe
- B. gcloud IAM zones describe
- C. gutil zones describe
- D. gcloud compute regions list
Q44.To create a custom role, a user must possess which role?
- A. IAM create
- B. compute.roles.create
- C. iam.roles.create
- D. Compute.roles.add
Q45.You have been asked to create a network with 1,000 IP addresses. In the interest of
minimizing unused IP addresses, which CIDR suffix would you use to create a network
with at least 1,000 addresses but no more than necessary?
- A. /20
- B. /22
- C. /28
- D. /32
Q46.A team of data scientists has asked for your help setting up an Apache Spark cluster.
You suggest they use a managed GCP service instead of managing a cluster themselves on
Compute Engine. The service they would use is:
- A. Cloud Dataproc
- B. Cloud Dataflow
- C. Cloud Hadoop
- D. BigQuery
Q47.You have created a web application that allows users to upload files to Cloud Storage.
When files are uploaded, you want to check the file size and update the user’s total storage
used in their account. A serverless option for performing this action on load is:
- A. Cloud Dataflow
- B. Cloud Dataproc
- C. Cloud Storage
- D. Cloud Functions
Q48.Your company has just started using GCP, and executives want to have a dedicated connection from your data center to the GCP to allow for large data transfers. Which networking service would you recommend?
- A. Google Cloud Carrier Internet Peering
- B. Google Cloud Interconnect – Dedicated
- C. Google Cloud Internet Peering
- D. Google Cloud DNS
Q49.You want to have GCP manage cryptographic keys, so you’ve decided to use Cloud Key
Management Services. Before you can start creating cryptographic keys, you must:
- A. Enable Google Cloud Key Management Service (KMS) API and set up billing
- B. Enable Google Cloud KMS API and create folders
- C. Create folders and set up billing
- D. Give all users grantable roles to create keys
Q50. In Kubernetes Engine, a node pool is:
- A. A subset of nodes across clusters
- B. A set of VMs managed outside of Kubernetes Engine
- C. A set of preemptible VMs
- D. A subset of node instances within a cluster that all have the same configuration
Q51. The GCP service for storing and managing Docker containers is:
- A. Cloud Source Repositories
- B. Cloud Build
- C. Container Registry
- D. Docker Repository
Q52. Code for Cloud Functions can be written in:
- A. Node.js and Python
- B. Node.js, Python, and Go
- C. Python and Go
- D. Python and C
Q53.Banking and the financial company have all IT estate on Google Cloud Platform with 8 different networks used by multiple departments with multiple accounts. Many Applications access data from other departments.
The security team thinks that there is data accessed by other applications /people. What will be your solution to identify if anyone is accessing as a quick solution?
- Ans: Flow Logs
Q54.Your company needs to run one compute engine all the time. If the instance if failed because of some reason, It should start one instance and keep running. This instance delivers HTTP traffic directly to the consumer and its low volume. You are tasked to identify a solution, What is your best solution?
- Ans: Create instance Group and Configure no of the instance to run to 1 and configure health check, if an application is running fine or not.
Q55.GCP allows you to configure an autoscaling policy compute engine. which of the following service will be able to help you to configure this?
- Ans: Instance Group
Q56.Compute Systems network performance depends on which of the following parameter?
- Ans: Number of vCPUs
Q57.Compute engines will need to have access to multiple users to have admin privileges for their own instances. What is a possible way to provide admin access?
- Ans: Set instance-level cloud IAM Policy and Apply it via Startup Script.
Q58.You want to provide SSH access to users based on the instance they are managing, What is the best way to do it?
- Ans: Add SSH Key to instance based on the user at startup of instances or manually using Console.
Q59. HTTP load balancer uses all of the following algorithms to distribute load across blackened instances except one.
- Ans: Memory Utilization.
Q60.Instances in instance groups are heavily utilized and do huge computing works. management applications want to know when the instance is live migrated to reduce further load when it’s being done so that its instance will not fail. How does the application know if the instance is getting live migrated?
- Ans: You can Capture live migration using event Compute, instances, MigrateOnHostMaintainance
Q61.When the compute instances in instance groups are not responding, the Instance group recreates the instance. You have your instance is running properly but not application serving HTTP traffic. How you can mark the application not running instance as bad and needs to recreate?
- Ans: Configure HTTP Health Check to check if the instance is running as well as the Application hosted in the instance is also running. You will need to open a firewall between health check /LB with the instance.
Q62.How can you add multiple applications in App Engine?
- Ans: Use a Different Service name than already present
Q63.The company has applications running in App Engine. Now the company realized that they have more customers in Europe than the USA where it’s hosted. The company decided to move the application to Europe, You as Cloud Engineer need to provide a solution to move the application. What is your quick solution?
- Ans: Create another project and create an App Engine with desired Europe location and deploy the application.
Q64.Which of the following is not scaling available in Google App Engine.
- Ans: On-demand
Q65. The financial company has a customer profile services application deployed in Google App Engine. The company wants to release a newer version of the application into production. service reliability engineering wants to release the application to some subscribers (5% only) and once they get confidence in the stability of the new version they want to release it to the customer. As a cloud engineer what you can do to satisfy their requirement as per Google’s best practices?
- Ans: Deploy the new version of the app and slowly move traffic 1 to 5% on the newer version, check application performance using multiple tools like Stack drive logging, Monitoring, Error Reporting and once newer version demonstrates stability as per their agreement move rest of the traffic to new version completely.
Q66.Google App engine flexible use container repository to store the image of container which gets deployed for application.
- Ans: True
Q67. Applications deployed in Kubernetes Engine need to store secrets(user/password of database or anything sensitive) somewhere, what options do you have?
- Ans: Secrets Object
Q68. You have the application deployed in the Kubernetes engine and you can scale it manually – currently, one instance of the pod is running, you want to run 3 instances of pods.
- Ans: kubectl scale deployment hello-web –replicas=3
Q69. Which of the following components in Kubernetes Engine you can manage different configurations of nodes together?
- Ans: Node Pools
Q70.In Kubernetes Engine, You can choose different PODs IP range than that of nodes.
- Ans: True
Q71.The financial company has the importance of each and every transaction. Currently, they are using Cloud SQL – MySQL instance with HA to store transaction data. The instance also has a read replica to serve read traffic. Business team concern about the point in time recovery in case of failure, What option will you configure to address their concern?
- Ans: Enable Binary Logging under HA configuration
Q72.Company is using Cloud SQL for many days now, Applications is now started facing performance issues on MYSQL They have scaled instance but there are many applications reads data and created problems when customer use those applications, What do you suggest to fix this problem?
- Ans: Create Read Replications for reading operations