Multilingual and Platform Independent Cloud Certification and Education App for AWS, Azure, Google Cloud

Cloud Edu Cert

Data Center Proxies - Data Collectors - Data Unblockers

The Cloud Education Certification App is an EduFlix App for AWS, Azure, Google Cloud Certification Prep [Android, iOS]

Technology is changing and is moving towards the cloud. The cloud will power most businesses in the coming years and is not taught in schools. How do we ensure that our kids and youth and ourselves are best prepared for this challenge?

Building mobile educational apps that work offline and on any device can help greatly in that sense.

Data Center Proxies - Data Collectors - Data Unblockers

The ability to tab on a button and learn the cloud fundamentals and take quizzes is a great opportunity to help our children and youth to boost their job prospects and be more productive at work.

The App covers the following certifications :
AWS Cloud Practitioner Exam Prep CCP CLF-C01, Azure Fundamentals AZ 900 Exam Prep, AWS Certified Solution Architect Associate SAA-C02 Exam Prep, AWS Certified Developer Associate DVA-C01 Exam Prep, Azure Administrator AZ 104 Exam Prep, Google Associate Cloud Engineer Exam Prep, Data Analytics for AWS DAS-C01, Machine Learning for AWS and Google, AWS Certified Security – Specialty (SCS-C01), AWS Certified Machine Learning – Specialty (MLS-C01), Google Cloud Professional Machine Learning Engineer and more… [Android, iOS]

Cloud Education Certification PRO: AWS Azure G.C.P
Cloud Education Certification PRO: AWS Azure G.C.P
Developer:
Etienne D. Noumen
Price:
$79.99

  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot
  • Cloud Education Certification PRO: AWS Azure G.C.P Screenshot

The App covers the following cloud categories:

AWS Technology, AWS Security and Compliance, AWS Cloud Concepts, AWS Billing and Pricing , AWS Design High Performing Architectures, AWS Design Cost Optimized Architectures, AWS Specify Secure Applications And Architectures, AWS Design Resilient Architecture, Development With AWS, AWS Deployment, AWS Security, AWS Monitoring, AWS Troubleshooting, AWS Refactoring, Azure Pricing and Support, Azure Cloud Concepts , Azure Identity, governance, and compliance, Azure Services , Implement and Manage Azure Storage, Deploy and Manage Azure Compute Resources, Configure and Manage Azure Networking Services, Monitor and Backup Azure Resources, GCP Plan and configure a cloud solution, GCP Deploy and implement a cloud solution, GCP Ensure successful operation of a cloud solution, GCP Configure access and security, GCP Setting up a cloud solution environment, AWS Incident Response, AWS Logging and Monitoring, AWS Infrastructure Security, AWS Identity and Access Management, AWS Data Protection, AWS Data Engineering, AWS Exploratory Data Analysis, AWS Modeling, AWS Machine Learning Implementation and Operations, GCP Frame ML problems, GCP Architect ML solutions, GCP Prepare and process data, GCP Develop ML models, GCP Automate & orchestrate ML pipelines, GCP Monitor, optimize, and maintain ML solutions, etc.. [Android, iOS]

Cloud Education and Certification

The App covers the following Cloud Services, Framework and technologies:

AWS: VPC, S3, DynamoDB, EC2, ECS, Lambda, API Gateway, CloudWatch, CloudTrail, Code Pipeline, Code Deploy, TCO Calculator, SES, EBS, ELB, AWS Autoscaling , RDS, Aurora, Route 53, Amazon CodeGuru, Amazon Bracket, AWS Billing and Pricing, Simply Monthly Calculator, cost calculator, Ec2 pricing on-demand, IAM, AWS Pricing, Pay As You Go, No Upfront Cost, Cost Explorer, AWS Organizations, Consolidated billing, Instance Scheduler, on-demand instances, Reserved instances, Spot Instances, CloudFront, Workspace, S3 storage classes, Regions, Availability Zones, Placement Groups, Amazon lightsail, Redshift, EC2 G4ad instances, DAAS, PAAS, IAAS, SAAS, NAAS, Machine Learning, Key Pairs, AWS CloudFormation, Amazon Macie, Amazon Textract, Glacier Deep Archive, 99.999999999% durability, AWS Codestar, Amazon Neptune, S3 Bucket, EMR, SNS, Desktop As A Service, Emazon EC2 for Mac, Aurora Postgres SQL, Kubernetes, Containers, Cluster.

Azure: Virtual Machines, Azure App Services, Azure Container Instances (ACI), Azure Kubernetes Service (AKS), and Windows Virtual Desktop, Virtual Networks, VPN Gateway, Virtual Network peering, and ExpressRoute, Container (Blob) Storage, Disk Storage, File Storage, and storage tiers, Cosmos DB, Azure SQL Database, Azure Database for MySQL, Azure Database for PostgreSQL, and SQL Managed Instance, Azure Marketplace, Azure consumption-based mode, management groups, resources and RG, Geographic distribution concepts such as Azure regions, region pairs, and AZ Internet of Things (IoT) Hub, IoT Central, and Azure Sphere, Azure Synapse Analytics, HDInsight, and Azure Databricks, Azure Machine Learning, Cognitive Services and Azure Bot Service, Serverless computing solutions that include Azure Functions and Logic Apps, Azure DevOps, GitHub, GitHub Actions, and Azure DevTest Labs, Azure Mobile, Azure Advisor, Azure Resource Manager (ARM) templates, Azure Security, Privacy and Workloads, General security and network security, Azure security features, Azure Security Centre, policy compliance, security alerts, secure score, and resource hygiene, Key Vault, Azure Sentinel, Azure Dedicated Hosts, Concept of defense in depth, NSG, Azure Firewall, Azure DDoS protection, Identity, governance, Conditional Access, Multi-Factor Authentication (MFA), and Single Sign-On (SSO),Azure Services, Core Azure architectural components, Management Groups, Azure Resource Manager,

Google Cloud Platform: Compute Engine, App Engine, BigQuery, Bigtable, Pub/Sub, flow logs, CORS, CLI, pod, Firebase, Cloud Run, Cloud Firestore, Cloud CDN, Cloud Storage, Persistent Disk, Kubernetes engine, Container registry, Cloud Load Balancing, Cloud Dataflow, gsutils, Cloud SQL,

Cloud Education Certification: Eduflix App for Cloud Education and Certification (AWS, Azure, Google Cloud) [Android, iOS]

Features:
– Practice exams
– 1000+ Q&A updated frequently.
– 3+ Practice exams per Certification
– Scorecard / Scoreboard to track your progress
– Quizzes with score tracking, progress bar, countdown timer.
– Can only see scoreboard after completing the quiz.
– FAQs for most popular Cloud services
– Cheat Sheets
– Flashcards
– works offline

Note and disclaimer: We are not affiliated with AWS, Azure, Microsoft or Google. The questions are put together based on the certification study guide and materials available online. The questions in this app should help you pass the exam but it is not guaranteed. We are not responsible for any exam you did not pass.

Important: To succeed with the real exam, do not memorize the answers in this app. It is very important that you understand why a question is right or wrong and the concepts behind it by carefully reading the reference documents in the answers.

Top 50 Google Certified Cloud Professional Architect Exam Questions and Answers Dumps

Azure Administrator AZ-104 Exam Questions and Answers Dumps

Data Center Proxies - Data Collectors - Data Unblockers

Google Certified Cloud Professional Architect is the top high paying certification in the world: Google Certified Professional Cloud Architect Average Salary – $175,761

The Google Certified Cloud Professional Architect Exam assesses your ability to:

  • Design and plan a cloud solution architecture
  • Manage and provision the cloud solution infrastructure
  • Design for security and compliance
  • Analyze and optimize technical and business processes
  • Manage implementations of cloud architecture
  • Ensure solution and operations reliability
  • Designing and planning a cloud solution architecture

The Google Certified Cloud Professional Architect covers the following topics:

Data Center Proxies - Data Collectors - Data Unblockers

Designing and planning a cloud solution architecture: 36%

This domain tests your ability to design a solution infrastructure that meets business and technical requirements and considers network, storage and compute resources. It will test your ability to create a migration plan, and that you can envision future solution improvements.

Managing and provisioning a solution Infrastructure: 20%

This domain will test your ability to configure network topologies, individual storage systems and design solutions using Google Cloud networking, storage and compute services.

Designing for security and compliance: 12%

This domain assesses your ability to design for security and compliance by considering IAM policies, separation of duties, encryption of data and that you can design your solutions while considering any compliance requirements such as those for healthcare and financial information.

Managing implementation: 10%

This domain tests your ability to advise development/operation team(s) to make sure you have successful deployment of your solution. It also tests yours ability to interact with Google Cloud using GCP SDK (gcloud, gsutil, and bq).

Ensuring solution and operations reliability: 6%

This domain tests your ability to run your solutions reliably in Google Cloud by building monitoring and logging solutions, quality control measures and by creating release management processes.

Analyzing and optimizing technical and business processes: 16%

This domain will test how you analyze and define technical processes, business processes and develop procedures to ensure resilience of your solutions in production.


Below are the Top 50 Google Certified Cloud Professional Architect Exam Questions and Answers Dumps: You will need to have the three case studies referred to in the exam open in separate tabs in order to complete the exam: Company A , Company B, Company C

Question 1:  Because you do not know every possible future use for the data Company A collects, you have decided to build a system that captures and stores all raw data in case you need it later. How can you most cost-effectively accomplish this goal?

 A. Have the vehicles in the field stream the data directly into BigQuery.

B. Have the vehicles in the field pass the data to Cloud Pub/Sub and dump it into a Cloud Dataproc cluster that stores data in Apache Hadoop Distributed File System (HDFS) on persistent disks.

C. Have the vehicles in the field continue to dump data via FTP, adjust the existing Linux machines, and use a collector to upload them into Cloud Dataproc HDFS for storage.

D. Have the vehicles in the field continue to dump data via FTP, and adjust the existing Linux machines to immediately upload it to Cloud Storage with gsutil.

ANSWER1:

D

Notes/References1:

D is correct because several load-balanced Compute Engine VMs would suffice to ingest 9 TB per day, and Cloud Storage is the cheapest per-byte storage offered by Google. Depending on the format, the data could be available via BigQuery immediately, or shortly after running through an ETL job. Thus, this solution meets business and technical requirements while optimizing for cost.

Reference: Streaming insertsApache Hadoop and Spark10 tips for building long running cluster using cloud dataproc


Associate Cloud Engineer Certification Prep PRO
Associate Cloud Engineer Certification Prep PRO
  • Associate Cloud Engineer Certification Prep PRO Screenshot
  • Associate Cloud Engineer Certification Prep PRO Screenshot
  • Associate Cloud Engineer Certification Prep PRO Screenshot
  • Associate Cloud Engineer Certification Prep PRO Screenshot
  • Associate Cloud Engineer Certification Prep PRO Screenshot
  • Associate Cloud Engineer Certification Prep PRO Screenshot
  • Associate Cloud Engineer Certification Prep PRO Screenshot
  • Associate Cloud Engineer Certification Prep PRO Screenshot
  • Associate Cloud Engineer Certification Prep PRO Screenshot
  • Associate Cloud Engineer Certification Prep PRO Screenshot



Question 2: Today, Company A maintenance workers receive interactive performance graphs for the last 24 hours (86,400 events) by plugging their maintenance tablets into the vehicle. The support group wants support technicians to view this data remotely to help troubleshoot problems. You want to minimize the latency of graph loads. How should you provide this functionality?

A. Execute queries against data stored in a Cloud SQL.

B. Execute queries against data indexed by vehicle_id.timestamp in Cloud Bigtable.

C. Execute queries against data stored on daily partitioned BigQuery tables.

D. Execute queries against BigQuery with data stored in Cloud Storage via BigQuery federation.

ANSWER2:

B

Notes/References2:

B is correct because Cloud Bigtable is optimized for time-series data. It is cost-efficient, highly available, and low-latency. It scales well. Best of all, it is a managed service that does not require significant operations work to keep running.

Reference: BigTables time series clusterBigQuery

Question 3: Your agricultural division is experimenting with fully autonomous vehicles. You want your architecture to promote strong security during vehicle operation. Which two architecture characteristics should you consider?

A. Use multiple connectivity subsystems for redundancy. 

B. Require IPv6 for connectivity to ensure a secure address space. 

C. Enclose the vehicle’s drive electronics in a Faraday cage to isolate chips.

D. Use a functional programming language to isolate code execution cycles.

Data Center Proxies - Data Collectors - Data Unblockers

E. Treat every microservice call between modules on the vehicle as untrusted.

F. Use a Trusted Platform Module (TPM) and verify firmware and binaries on boot.

ANSWER3:

E and F

Notes/References3:

E is correct because this improves system security by making it more resistant to hacking, especially through man-in-the-middle attacks between modules.

F is correct because this improves system security by making it more resistant to hacking, especially rootkits or other kinds of corruption by malicious actors.

Reference 3: Trusted Platform Module

Question 4: For this question, refer to the Company A case study.

Which of Company A’s legacy enterprise processes will experience significant change as a result of increased Google Cloud Platform adoption?

A. OpEx/CapEx allocation, LAN change management, capacity planning

B. Capacity planning, TCO calculations, OpEx/CapEx allocation 

C. Capacity planning, utilization measurement, data center expansion

D. Data center expansion, TCO calculations, utilization measurement

ANSWER4:

B

Notes/References4:

B is correct because all of these tasks are big changes when moving to the cloud. Capacity planning for cloud is different than for on-premises data centers; TCO calculations are adjusted because Company A is using services, not leasing/buying servers; OpEx/CapEx allocation is adjusted as services are consumed vs. using capital expenditures.

Reference: Cloud Economics




Question 5: For this question, refer to the Company A case study.

You analyzed Company A’s business requirement to reduce downtime and found that they can achieve a majority of time saving by reducing customers’ wait time for parts. You decided to focus on reduction of the 3 weeks’ aggregate reporting time. Which modifications to the company’s processes should you recommend?

A. Migrate from CSV to binary format, migrate from FTP to SFTP transport, and develop machine learning analysis of metrics.

B. Migrate from FTP to streaming transport, migrate from CSV to binary format, and develop machine learning analysis of metrics.

C. Increase fleet cellular connectivity to 80%, migrate from FTP to streaming transport, and develop machine learning analysis of metrics.

D. Migrate from FTP to SFTP transport, develop machine learning analysis of metrics, and increase dealer local inventory by a fixed factor.

ANSWER5:

C

Notes/References5:

C is correct because using cellular connectivity will greatly improve the freshness of data used for analysis from where it is now, collected when the machines are in for maintenance. Streaming transport instead of periodic FTP will tighten the feedback loop even more. Machine learning is ideal for predictive maintenance workloads.

Question 6: Your company wants to deploy several microservices to help their system handle elastic loads. Each microservice uses a different version of software libraries. You want to enable their developers to keep their development environment in sync with the various production services. Which technology should you choose?

A. RPM/DEB

B. Containers 

C. Chef/Puppet

D. Virtual machines

ANSWER6:

B

Notes/References6:

B is correct because using containers for development, test, and production deployments abstracts away system OS environments, so that a single host OS image can be used for all environments. Changes that are made during development are captured using a copy-on-write filesystem, and teams can easily publish new versions of the microservices in a repository.

Question 7: Your company wants to track whether someone is present in a meeting room reserved for a scheduled meeting. There are 1000 meeting rooms across 5 offices on 3 continents. Each room is equipped with a motion sensor that reports its status every second. You want to support the data upload and collection needs of this sensor network. The receiving infrastructure needs to account for the possibility that the devices may have inconsistent connectivity. Which solution should you design?

A. Have each device create a persistent connection to a Compute Engine instance and write messages to a custom application.

B. Have devices poll for connectivity to Cloud SQL and insert the latest messages on a regular interval to a device specific table. 

C. Have devices poll for connectivity to Cloud Pub/Sub and publish the latest messages on a regular interval to a shared topic for all devices.

D. Have devices create a persistent connection to an App Engine application fronted by Cloud Endpoints, which ingest messages and write them to Cloud Datastore.

ANSWER7:

C

Notes/References7:

C is correct because Cloud Pub/Sub can handle the frequency of this data, and consumers of the data can pull from the shared topic for further processing.




Question 8: Your company wants to try out the cloud with low risk. They want to archive approximately 100 TB of their log data to the cloud and test the analytics features available to them there, while also retaining that data as a long-term disaster recovery backup. Which two steps should they take?

A. Load logs into BigQuery. 

B. Load logs into Cloud SQL.

C. Import logs into Stackdriver. 

D. Insert logs into Cloud Bigtable.

E. Upload log files into Cloud Storage.

ANSWER8:

A and E

Notes/References8:

A is correct because BigQuery is the fully managed cloud data warehouse for analytics and supports the analytics requirement.

E is correct because Cloud Storage provides the Coldline storage class to support long-term storage with infrequent access, which would support the long-term disaster recovery backup requirement.

References: BigQueryStackDriverBigTableStorage Class: ColdLine




Question 9: You set up an autoscaling instance group to serve web traffic for an upcoming launch. After configuring the instance group as a backend service to an HTTP(S) load balancer, you notice that virtual machine (VM) instances are being terminated and re-launched every minute. The instances do not have a public IP address. You have verified that the appropriate web response is coming from each instance using the curl command. You want to ensure that the backend is configured correctly. What should you do?

A. Ensure that a firewall rule exists to allow source traffic on HTTP/HTTPS to reach the load balancer. 

B. Assign a public IP to each instance, and configure a firewall rule to allow the load balancer to reach the instance public IP.

C. Ensure that a firewall rule exists to allow load balancer health checks to reach the instances in the instance group.

D. Create a tag on each instance with the name of the load balancer. Configure a firewall rule with the name of the load balancer as the source and the instance tag as the destination.

ANSWER9:

C

Notes/References9:

C is correct because health check failures lead to a VM being marked unhealthy and can result in termination if the health check continues to fail. Because you have already verified that the instances are functioning properly, the next step would be to determine why the health check is continuously failing.

Reference: Load balancingLoad Balancing Health Checking

Question 10: Your organization has a 3-tier web application deployed in the same network on Google Cloud Platform. Each tier (web, API, and database) scales independently of the others. Network traffic should flow through the web to the API tier, and then on to the database tier. Traffic should not flow between the web and the database tier. How should you configure the network?

A. Add each tier to a different subnetwork.

B. Set up software-based firewalls on individual VMs. 

C. Add tags to each tier and set up routes to allow the desired traffic flow.

D. Add tags to each tier and set up firewall rules to allow the desired traffic flow.

ANSWER10:

D

Notes/References10:

D is correct because as instances scale, they will all have the same tag to identify the tier. These tags can then be leveraged in firewall rules to allow and restrict traffic as required, because tags can be used for both the target and source.

Reference: Using VPCRoutesAdd Remove Network




Question 11: Your organization has 5 TB of private data on premises. You need to migrate the data to Cloud Storage. You want to maximize the data transfer speed. How should you migrate the data?

A. Use gsutil.

B. Use gcloud.

C. Use GCS REST API. 

D. Use Storage Transfer Service.

ANSWER11:

A

Notes/References11:

A is correct because gsutil gives you access to write data to Cloud Storage.

Reference: gsutilsgcloud sdkcloud storage json apiuploading objectsstorage transfer

Question 12: You are designing a mobile chat application. You want to ensure that people cannot spoof chat messages by proving that a message was sent by a specific user. What should you do?

A. Encrypt the message client-side using block-based encryption with a shared key.

B. Tag messages client-side with the originating user identifier and the destination user.

C. Use a trusted certificate authority to enable SSL connectivity between the client application and the server. 

D. Use public key infrastructure (PKI) to encrypt the message client-side using the originating user’s private key.

ANSWER12:

D

Notes/References12:

D is correct because PKI requires that both the server and the client have signed certificates, validating both the client and the server.

Question 13: You are designing a large distributed application with 30 microservices. Each of your distributed microservices needs to connect to a database backend. You want to store the credentials securely. Where should you store the credentials?

A. In the source code

B. In an environment variable 

C. In a key management system

D. In a config file that has restricted access through ACLs

ANSWER13:

C

Notes/References13:




Question 14: For this question, refer to the Company B case study.

Company B wants to set up a real-time analytics platform for their new game. The new platform must meet their technical requirements. Which combination of Google technologies will meet all of their requirements?

A. Kubernetes Engine, Cloud Pub/Sub, and Cloud SQL

B. Cloud Dataflow, Cloud Storage, Cloud Pub/Sub, and BigQuery 

C. Cloud SQL, Cloud Storage, Cloud Pub/Sub, and Cloud Dataflow

D. Cloud Pub/Sub, Compute Engine, Cloud Storage, and Cloud Dataproc

ANSWER14:

B

Notes/References14:

B is correct because:
Cloud Dataflow dynamically scales up or down, can process data in real time, and is ideal for processing data that arrives late using Beam windows and triggers.
Cloud Storage can be the landing space for files that are regularly uploaded by users’ mobile devices.
Cloud Pub/Sub can ingest the streaming data from the mobile users.
BigQuery can query more than 10 TB of historical data.

References: GCP QuotasBeam Apache WindowingBeam Apache TriggersBigQuery External Data SolutionsApache Hive on Cloud Dataproc

Question 15: For this question, refer to the Company B case study.

Company B has deployed their new backend on Google Cloud Platform (GCP). You want to create a thorough testing process for new versions of the backend before they are released to the public. You want the testing environment to scale in an economical way. How should you design the process?A. Create a scalable environment in GCP for simulating production load.B. Use the existing infrastructure to test the GCP-based backend at scale. C. Build stress tests into each component of your application and use resources from the already deployed production backend to simulate load.D. Create a set of static environments in GCP to test different levels of load—for example, high, medium, and low.

ANSWER15:

A

Notes/References15:

A is correct because simulating production load in GCP can scale in an economical way.

Reference: Load Testing iot using gcp and locustDistributed Load Testing Using Kubernetes

Question 16: For this question, refer to the Company B case study.

Company B wants to set up a continuous delivery pipeline. Their architecture includes many small services that they want to be able to update and roll back quickly. Company B has the following requirements:

  • Services are deployed redundantly across multiple regions in the US and Europe
  • Only frontend services are exposed on the public internet.
  • They can reserve a single frontend IP for their fleet of services.
  • Deployment artifacts are immutable

Which set of products should they use?

A. Cloud Storage, Cloud Dataflow, Compute Engine

B. Cloud Storage, App Engine, Cloud Load Balancing

C. Container Registry, Google Kubernetes Engine, Cloud Load Balancing

D. Cloud Functions, Cloud Pub/Sub, Cloud Deployment Manager

ANSWER16:

C

Notes/References16:

C is correct because:
Google Kubernetes Engine is ideal for deploying small services that can be updated and rolled back quickly. It is a best practice to manage services using immutable containers.
Cloud Load Balancing supports globally distributed services across multiple regions. It provides a single global IP address that can be used in DNS records. Using URL Maps, the requests can be routed to only the services that Company B wants to expose.
Container Registry is a single place for a team to manage Docker images for the services.

References: Load Balancing https – load balancing overview GCP lb global forwarding rulesreserve static external ip addressbest practice for operating containerscontainer registrydataflowcalling https

Question 17: Your customer is moving their corporate applications to Google Cloud Platform. The security team wants detailed visibility of all resources in the organization. You use Resource Manager to set yourself up as the org admin. What Cloud Identity and Access Management (Cloud IAM) roles should you give to the security team?

A. Org viewer, Project owner

B. Org viewer, Project viewer 

C. Org admin, Project browser

D. Project owner, Network admin

ANSWER17:

B

Notes/References17:

B is correct because:
Org viewer grants the security team permissions to view the organization's display name.
Project viewer grants the security team permissions to see the resources within projects.

Reference: GCP Resource Manager – User Roles

Question 18: To reduce costs, the Director of Engineering has required all developers to move their development infrastructure resources from on-premises virtual machines (VMs) to Google Cloud Platform. These resources go through multiple start/stop events during the day and require state to persist. You have been asked to design the process of running a development environment in Google Cloud while providing cost visibility to the finance department. Which two steps should you take?

A. Use persistent disks to store the state. Start and stop the VM as needed. 

B. Use the –auto-delete flag on all persistent disks before stopping the VM. 

C. Apply VM CPU utilization label and include it in the BigQuery billing export.

D. Use BigQuery billing export and labels to relate cost to groups. 

E. Store all state in local SSD, snapshot the persistent disks, and terminate the VM.F. Store all state in Cloud Storage, snapshot the persistent disks, and terminate the VM.

ANSWER18:

A and D

Notes/References18:

A is correct because persistent disks will not be deleted when an instance is stopped.

D is correct because exporting daily usage and cost estimates automatically throughout the day to a BigQuery dataset is a good way of providing visibility to the finance department. Labels can then be used to group the costs based on team or cost center.

References: GCP instances life cycleGCP instances set disk auto deleteGCP Local Data PersistanceGCP export data BigQueryGCP Creating Managing Labels




Question 19: Your company has decided to make a major revision of their API in order to create better experiences for their developers. They need to keep the old version of the API available and deployable, while allowing new customers and testers to try out the new API. They want to keep the same SSL and DNS records in place to serve both APIs. What should they do?

A. Configure a new load balancer for the new version of the API.

B. Reconfigure old clients to use a new endpoint for the new API. 

C. Have the old API forward traffic to the new API based on the path.

D. Use separate backend services for each API path behind the load balancer.

ANSWER19:

D

Notes/References19:

D is correct because an HTTP(S) load balancer can direct traffic reaching a single IP to different backends based on the incoming URL.

References: load balancing httpsload balancing backendGCP lb global forwarding rules

Question 20: The database administration team has asked you to help them improve the performance of their new database server running on Compute Engine. The database is used for importing and normalizing the company’s performance statistics. It is built with MySQL running on Debian Linux. They have an n1-standard-8 virtual machine with 80 GB of SSD zonal persistent disk. What should they change to get better performance from this system in a cost-effective manner?

A. Increase the virtual machine’s memory to 64 GB.

B. Create a new virtual machine running PostgreSQL. 

C. Dynamically resize the SSD persistent disk to 500 GB.

D. Migrate their performance metrics warehouse to BigQuery.

ANSWER20:

C

Notes/References20:

C is correct because persistent disk performance is based on the total persistent disk capacity attached to an instance and the number of vCPUs that the instance has. Incrementing the persistent disk capacity will increment its throughput and IOPS, which in turn improve the performance of MySQL.

References: GCP compute disks pdsspecsGCP Compute Disks Performances




Question 21: You need to ensure low-latency global access to data stored in a regional GCS bucket. Data access is uniform across many objects and relatively high. What should you do to address the latency concerns?

A. Use Google’s Cloud CDN.

B. Use Premium Tier routing and Cloud Functions to accelerate access at the edges.

C. Do nothing.

D. Use global BigTable storage.

E. Use a global Cloud Spanner instance.

F. Migrate the data to a new multi-regional GCS bucket.

G. Change the storage class to multi-regional.

ANSWER21:

A

Notes/References21:

Cloud Functions cannot be used to affect GCS data access, so that option is simply wrong. BigTable does not have any “global” mode, so that option is wrong, too. Cloud Spanner is not a good replacement for GCS data: the data use cases are different enough that we can assume it would probably not be a good fit. You cannot change a bucket’s location after it has been created–not via the storage class nor any other way; you would have to migrate the data to a new bucket. Google’s Cloud CDN is very easy to turn on, but it does only work for data that comes from within GCP and only if the objects are being accessed frequently enough. 

Reference: Google Cloud Storage : What bucket class for the best performance?

Question 22: You are building a sign-up app for your local neighbourhood barbeque party and you would like to quickly throw together a low-cost application that tracks who will bring what. Which of the following options should you choose?

A. Python, Flask, App Engine Standard

B. Ruby, Nginx, GKE

C. HTML, CSS, Cloud Storage

D. Node.js, Express, Cloud Functions

E. Rust, Rocket, App Engine Flex

F. Perl, CGI, GCE

ANSWER22:

A

Notes/References22:

The Cloud Storage option doesn’t offer any way to coordinate the guest data. App Engine Flex would cost much more to run when no one is on the sign-up site. Cloud Functions could handle processing some API calls, but it would be more work to set up and that option doesn’t mention anything about storage. GKE is way overkill for such a small and simple application. Running Perl CGI scripts on GCE would also cost more than it needs (and probably make you very sad). App Engine Standard makes it super-easy to stand up a Python Flask app and includes easy data storage options, too. 

Reference: Building a Python 3.7 App on App Engine

Question 23: Your company has decided to migrate your AWS DynamoDB database to a multi-regional Cloud Spanner instance and you are designing the system to transfer and load all the data to synchronize the DBs and eventually allow for a quick cut-over. A member of your team has some previous experience working with Apache Hadoop. Which of the following options will you choose for the streamed updates that follow the initial import?

A. The DynamoDB table change is captured by Cloud Pub/Sub and written to Cloud Dataproc for processing into a Spanner-compatible format.

B. The DynamoDB table change is captured by Cloud Pub/Sub and written to Cloud Dataflow for processing into a Spanner-compatible format.

C. Changes to the DynamoDB table are captured by DynamoDB Streams. A Lambda function triggered by the stream writes the change to Cloud Pub/Sub. Cloud Dataflow processes the data from Cloud Pub/Sub and writes it to Cloud Spanner.

D. The DynamoDB table is rescanned by a GCE instance and written to a Cloud Storage bucket. Cloud Dataproc processes the data from Cloud Storage and writes it to Cloud Spanner.

E. The DynamoDB table is rescanned by an EC2 instance and written to an S3 bucket. Storage Transfer Service moves the data from S3 to a Cloud Storage bucket. Cloud Dataflow processes the data from Cloud Storage and writes it to Cloud Spanner.

ANSWER23:

C

Notes/References23:

Rescanning the DynamoDB table is not an appropriate approach to tracking data changes to keep the GCP-side of this in synch. The fact that someone on your team has previous Hadoop experience is not a good enough reason to choose Cloud Dataproc; that’s a red herring. The options purporting to connect Cloud Pub/Sub directly to the DynamoDB table won’t work because there is no such functionality. 

References: Cloud Solutions Architecture Reference


Question 24: Your client is a manufacturing company and they have informed you that they will be pausing all normal business activities during a five-week summer holiday period. They normally employ thousands of workers who constantly connect to their internal systems for day-to-day manufacturing data such as blueprints and machine imaging, but during this period the few on-site staff will primarily be re-tooling the factory for the next year’s production runs and will not be performing any manufacturing tasks that need to access these cloud-based systems. When the bulk of the staff return, they will primarily work on the new models but may spend about 20% of their time working with models from previous years. The company has asked you to reduce their GCP costs during this time, so which of the following options will you suggest?

A. Pause all Cloud Functions via the UI and unpause them when work starts back up.

B. Disable all Cloud Functions via the command line and re-enable them when work starts back up.

C. Delete all Cloud Functions and recreate them when work starts back up.

D. Convert all Cloud Functions to run as App Engine Standard applications during the break.

E. None of these options is a good suggestion.

ANSWER24:

E

Notes/References24:

Cloud Functions scale themselves down to zero when they’re not being used. There is no need to do anything with them.

Question 25: You need a place to store images before updating them by file-based render farm software running on a cluster of machines. Which of the following options will you choose?

A. Container Registry

B. Cloud Storage

C. Cloud Filestore

D. Persistent Disk

ANSWER25:

C

Notes/References25:

There are several different kinds of “images” that you might need to consider–maybe they are normal picture-image files, maybe they are Docker container images, maybe VM or disk images, or maybe something else. In this question, “images” refers to visual images, thus eliminating CI/CD products like Container Registry. Compute Engine is not a storage product and should be eliminated. The term “file-based” software means that it is unlikely to work well with object-based storage like Cloud Storage (or any of its storage classes). Persistent Disk cannot offer shared access across a cluster of machines when writes are involved; it only handles multiple readers. However, Cloud Filestore is made to provide shared, file-based storage for a cluster of machines as described in the question. 

Reference: Cloud Filestore | Google Cloud

Question 26: Your company has decided to migrate your AWS DynamoDB database to a multi-regional Cloud Spanner instance and you are designing the system to transfer and load all the data to synchronize the DBs and eventually allow for a quick cut-over. A member of your team has some previous experience working with Apache Hadoop. Which of the following options will you choose for the initial data import?

A. The DynamoDB table is scanned by an EC2 instance and written to an S3 bucket. Storage Transfer Service moves the data from S3 to a Cloud Storage bucket. Cloud Dataflow processes the data from Cloud Storage and writes it to Cloud Spanner.

B. The DynamoDB table data is captured by DynamoDB Streams. A Lambda function triggered by the stream writes the data to Cloud Pub/Sub. Cloud Dataflow processes the data from Cloud Pub/Sub and writes it to Cloud Spanner.

C. The DynamoDB table data is captured by Cloud Pub/Sub and written to Cloud Dataproc for processing into a Spanner-compatible format.

D. The DynamoDB table is scanned by a GCE instance and written to a Cloud Storage bucket. Cloud Dataproc processes the data from Cloud Storage and writes it to Cloud Spanner.

ANSWER26:

A

Notes/References26:

The same data processing will have to happen for both the initial (batch) data load and the incremental (streamed) data changes that follow it. So if the solution built to handle the initial batch doesn't also work for the stream that follows it, then the processing code would have to be written twice. A Professional Cloud Architect should recognize this project-level issue and not over-focus on the (batch) portion called out in this particular question. This is why you don’t want to choose Cloud Dataproc. Instead, Cloud Dataflow will handle both the initial batch load and also the subsequent streamed data. The fact that someone on your team has previous Hadoop experience is not a good enough reason to choose Cloud Dataproc; that’s a red herring. The DynamoDB streams option would be great for the db synchronization that follows, but it can’t handle the initial data load because DynamoDB Streams only fire for data changes. The option purporting to connect Cloud Pub/Sub directly to the DynamoDB table won’t work because there is no such functionality. 

Reference: Cloud Solutions Architecture Reference


Question 27: You need a managed service to handle logging data coming from applications running in GKE and App Engine Standard. Which option should you choose?

A. Cloud Storage

B. Logstash

C. Cloud Monitoring

D. Cloud Logging

E. BigQuery

F. BigTable

ANSWER27:

D

Notes/References27:

Cloud Monitoring is made to handle metrics, not logs. Logstash is not a managed service. And while you could store application logs in almost any storage service, the Cloud Logging service–aka Stackdriver Logging–is purpose-built to accept and process application logs from many different sources. Oh, and you should also be comfortable dealing with products and services by names other than their current official ones. For example, “GKE” used to be called “Container Engine”, “Cloud Build” used to be “Container Builder”, the “GCP Marketplace” used to be called “Cloud Launcher”, and so on. 

Reference: Cloud Logging | Google Cloud

Question 28: You need a place to store images before serving them from AppEngine Standard. Which of the following options will you choose?

A. Compute Engine

B. Cloud Filestore

C. Cloud Storage

D. Persistent Disk

E. Container Registry

F. Cloud Source Repositories

G. Cloud Build

H. Nearline

ANSWER28:

C

Notes/References28:

There are several different kinds of “images” that you might need to consider–maybe they are normal picture-image files, maybe they are Docker container images, maybe VM or disk images, or maybe something else. In this question, “images” refers to picture files, because that’s something that you would serve from a web server product like AppEngine Standard, so we eliminate Cloud Build (which isn’t actually for storage, at all) and the other two CI/CD products: Cloud Source Repositories and Container Registry. You definitely could store image files on Cloud Filestore or Persistent Disk, but you can’t hook those up to AppEngine Standard, so those options need to be eliminated, too. The only options left are both types of Cloud Storage, but since “Cloud Storage” sits next to “Coldline” as an option, we can confidently infer that the former refers to the “Standard” storage class. Since the question implies that these images will be served by AppEngine Standard, we would prefer to use the Standard storage class over the Coldline one–so there’s our answer. 

Reference: The App Engine Standard Environment Cloud Storage: Object Storage | Google Cloud Storage classes | Cloud Storage | Google Cloud

Question 29: You need to ensure low-latency global access to data stored in a multi-regional GCS bucket. Data access is uniform across many objects and relatively low. What should you do to address the latency concerns?

A. Use a global Cloud Spanner instance.

B. Change the storage class to multi-regional.

C. Use Google’s Cloud CDN.

D. Migrate the data to a new regional GCS bucket.

E. Do nothing.

F. Use global BigTable storage.

ANSWER29:

E

Notes/References29:

Cloud Functions cannot be used to affect GCS data access, so that option is simply wrong. BigTable does not have any “global” mode, so that option is wrong, too. Cloud Spanner is not a good replacement for GCS data: the data use cases are different enough that we can assume it would probably not be a good fit. You cannot change a bucket’s location after it has been created–not via the storage class nor any other way; you would have to migrate the data to a new bucket. But migrating the data to a regional bucket only helps when the data access will primarily be from that region. Google’s Cloud CDN is very easy to turn on, but it does only work for data that comes from within GCP and only if the objects are being accessed frequently enough to get cached based on previous requests. Because the access per object is so low, Cloud CDN won’t really help. This then brings us back to the question. Now, it may seem implied, but the question does not specifically state that there is currently a problem with latency, only that you need to ensure low latency–and we are already using what would be the best fit for this situation: a multi-regional CS bucket. 

Reference: Google Cloud Storage : What bucket class for the best performance?


Question 30: You need to ensure low-latency GCP access to a volume of historical data that is currently stored in an S3 bucket. Data access is uniform across many objects and relatively high. What should you do to address the latency concerns?

A. Use Premium Tier routing and Cloud Functions to accelerate access at the edges.

B. Use Google’s Cloud CDN.

C. Use global BigTable storage.

D. Do nothing.

E. Migrate the data to a new multi-regional GCS bucket.

F. Use a global Cloud Spanner instance.

ANSWER30:

E

Notes/References30:

Cloud Functions cannot be used to affect GCS data access, so that option is simply wrong. BigTable does not have any “global” mode, so that option is wrong, too. Cloud Spanner is not a good replacement for GCS data: the data use cases are different enough that we can assume it would probably not be a good fit–and it would likely be unnecessarily expensive. You cannot change a bucket’s location after it has been created–not via the storage class nor any other way; you would have to migrate the data to a new bucket. Google’s Cloud CDN is very easy to turn on, but it does only work for data that comes from within GCP and only if the objects are being accessed frequently enough. So even if you would want to use Cloud CDN, you have to migrate the data into a GCS bucket first, so that’s a better option. 

Reference: Google Cloud Storage : What bucket class for the best performance?

Question 31: You are lifting and shifting into GCP a system that uses a subnet-based security model. It has frontend and backend tiers and will be deployed in three regions. How many subnets will you need?

A. Six

B. One

C. Three

D. Four

E. Two

F. Nine

ANSWER31:

A

Notes/References31:

A single subnet spans and can be used across all zones in a single region, but you will need different subnets in different regions. Also, to implement subnet-level network security, you need to separate each tier into its own subnet. In this case, you have two tiers which will each need their own subnet in each of the three regions in which you will deploy this system. 

Reference: VPC network overview | Google Cloud Best practices and reference architectures for VPC design | Solutions


Question 32: You need a place to produce images before deploying them to AppEngine Flex. Which of the following options will you choose?

A. Container Registry

B. Cloud Storage

C. Persistent Disk

D. Nearline

E. Cloud Source Repositories

F. Cloud Build

G. Cloud Filestore

H. Compute Engine

ANSWER32:

F

Notes/References32:

There are several different kinds of “images” that you might need to consider–maybe they are normal picture-image files, maybe they are Docker container images, maybe VM or disk images, or maybe something else. In this question, “deploying [these images] to AppEngine Flex” lets us know that we are dealing with Docker container images, and thus although they would likely be stored in the Container Registry, after being built, this question asks us where that building might happen, which is Cloud Build. Cloud Build, which used to be called Container Builder, is ideal for building container images–though it can also be used to build almost any artifacts, really. You could also do this on Compute Engine, but that option requires much more work to manage and is therefore worse. 

Reference: Google App Engine flexible environment docs | Google Cloud Container Registry | Google Cloud

Question 33: You are lifting and shifting into GCP a system that uses a subnet-based security model. It has frontend, app, and data tiers and will be deployed in three regions. How many subnets will you need?

A. Two

B. One

C. Three

D. Nine

E. Four

F. Six

ANSWER33:

D

Notes/References33:

A single subnet spans and can be used across all zones in a single region, but you will need different subnets in different regions. Also, to implement subnet-level network security, you need to separate each tier into its own subnet. In this case, you have three tiers which will each need their own subnet in each of the three regions in which you will deploy this system. 

Reference: VPC network overview | Google Cloud Best practices and reference architectures for VPC design | Solutions

Question 34: You need a place to store images in case any of them are needed as evidence for a tax audit over the next seven years. Which of the following options will you choose?

A. Cloud Filestore

B. Coldline

C. Nearline

D. Persistent Disk

E. Cloud Source Repositories

F. Cloud Storage

G. Container Registry

ANSWER34:

B

Notes/References34:

There are several different kinds of “images” that you might need to consider–maybe they are normal picture-image files, maybe they are Docker container images, maybe VM or disk images, or maybe something else. In this question, “images” probably refers to picture files, and so Cloud Storage seems like an interesting option. But even still, when “Cloud Storage” is used without any qualifier, it generally refers to the “Standard” storage class, and this question also offers other storage classes as response options. Because the images in this scenario are unlikely to be used more than once a year (we can assume that taxes are filed annually and there’s less than 100% chance of being audited), the right storage class is Coldline. 

Reference: Cloud Storage: Object Storage | Google Cloud Storage classes | Cloud Storage | Google Cloud


Question 35: You need a place to store images before deploying them to AppEngine Flex. Which of the following options will you choose?

A. Container Registry

B. Cloud Filestore

C. Cloud Source Repositories

D. Persistent Disk

E. Cloud Storage

F. Code Build

G. Nearline

ANSWER35:

A

Notes/References35:

Compute Engine is not a storage product and should be eliminated. There are several different kinds of “images” that you might need to consider–maybe they are normal picture-image files, maybe they are Docker container images, maybe VM or disk images, or maybe something else. In this question, “deploying [these images] to AppEngine Flex” lets us know that we are dealing with Docker container images, and thus they would likely have been stored in the Container Registry. 

Reference: Google App Engine flexible environment docs | Google Cloud Container Registry | Google Cloud

Question 36: You are configuring a SaaS security application that updates your network’s allowed traffic configuration to adhere to internal policies. How should you set this up?

A. Install the application on a new appropriately-sized GCE instance running in your host VPC, and apply a read-only service account to it.

B. Create a new service account for the app to use and grant it the compute.networkViewer role on the production VPC.

C. Create a new service account for the app to use and grant it the compute.securityAdmin role on the production VPC.

D. Run the application as a container in your system’s staging GKE cluster and grant it access to a read-only service account.

E. Install the application on a new appropriately-sized GCE instance running in your host VPC, and let it use the default service account.

ANSWER36:

C

Notes/References36:

You do not install a Software-as-a-Service application yourself; instead, it runs on the vendor's own hardware and you configure it for external access. Service accounts are great for this, as they can be used externally and you maintain full control over them (disabling them, rotating their keys, etc.). The principle of least privilege dictates that you should not give any application more ability than it needs, but this app does need to make changes, so you'll need to grant securityAdmin, not networkViewer. 

Reference: VPC network overview | Google Cloud Best practices and reference architectures for VPC design | Solutions Understanding roles | Cloud IAM Documentation | Google Cloud

Question 37: You are lifting and shifting into GCP a system that uses a subnet-based security model. It has frontend and backend tiers and will be deployed across three zones. How many subnets will you need?

A. One

B. Six

C. Four

D. Three

E. Nine

ANSWER37:

F

Notes/References37:

A single subnet spans and can be used across all zones in a given region. But to implement subnet-level network security, you need to separate each tier into its own subnet. In this case, you have two tiers, so you only need two subnets. 

Reference: VPC network overview | Google Cloud Best practices and reference architectures for VPC design | Solutions


Question 38: You have been tasked with setting up a system to comply with corporate standards for container image approvals. Which of the following is your best choice for this project?

A. Binary Authorization

B. Cloud IAM

C. Security Key Enforcement

D. Cloud SCC

E. Cloud KMS

ANSWER38:

A

Notes/References38:

Cloud KMS is Google's product for managing encryption keys. Security Key Enforcement is about making sure that people's accounts do not get taken over by attackers, not about managing encryption keys. Cloud IAM is about managing what identities (both humans and services) can access in GCP. Cloud DLP–or Data Loss Prevention–is for preventing data loss by scanning for and redacting sensitive information. Cloud SCC–the Security Command Center–centralizes security information so you can manage it all in one place. Binary Authorization is about making sure that only properly-validated containers can run in your environments. 

Reference: Cloud Key Management Service | Google Cloud Cloud IAM | Google Cloud Cloud Data Loss Prevention | Google Cloud Security Command Center | Google Cloud Binary Authorization | Google Cloud Security Key Enforcement – 2FA

Question 39: For this question, refer to the Company B‘s case study. Which of the following are most likely to impact the operations of Company B’s game backend and analytics systems?

A. PCI

B. PII

C. SOX

D. GDPR

E. HIPAA

ANSWER39:

B and D

Notes/References39:

There is no patient/health information, so HIPAA does not apply. It would be a very bad idea to put payment card information directly into these systems, so we should assume they’ve not done that–therefore the Payment Card Industry (PCI) standards/regulations should not affect normal operation of these systems. Besides, it’s entirely likely that they never deal with payments directly, anyway–choosing to offload that to the relevant app stores for each mobile platform. Sarbanes-Oxley (SOX) is about proper management of financial records for publicly traded companies and should therefore not apply to these systems. However, these systems are likely to contain some Personally-Identifying Information (PII) about the users who may reside in the European Union and therefore the EU’s General Data Protection Regulations (GDPR) will apply and may require ongoing operations to comply with the “Right to be Forgotten/Erased”. 

Reference: Sarbanes–Oxley Act – Wikipedia Payment Card Industry Data Security Standard – Wikipedia Personal data – Wikipedia Personal data – Wikipedia

Question 40: Your new client has advised you that their organization falls within the scope of HIPAA. What can you infer about their information systems?

A. Their customers located in the EU may require them to delete their user data and provide evidence of such.

B. They will also need to pass a SOX audit.

C. They handle money-linked information.

D. Their system deals with medical information.

ANSWER40:

D

Notes/References40:

SOX stands for Sarbanes Oxley and is US regulation governing financial reporting for publicly-traded companies. HIPAA–the Health Insurance Portability and Accountability Act of 1996–is US regulation aimed at safeguarding individuals' (i.e. patients’) health information. PCI is the Payment Card Industry, and they have Data Security Standards (DSS) that must be adhered to by systems handling payment information of any of their member brands (which include Visa, Mastercard, and several others). 

Reference: Cloud Compliance & Regulations Resources | Google Cloud

Question 41: Your new client has advised you that their organization needs to pass audits by ISO and PCI. What can you infer about their information systems?

A. They handle money-linked information.

B. Their customers located in the EU may require them to delete their user data and provide evidence of such.

C. Their system deals with medical information.

D. They will also need to pass a SOX audit.

ANSWER42:

A

Notes/References42:

SOX stands for Sarbanes Oxley and is US regulation governing financial reporting for publicly-traded companies. HIPAA–the Health Insurance Portability and Accountability Act of 1996–is US regulation aimed at safeguarding individuals' (i.e. patients’) health information. PCI is the Payment Card Industry, and they have Data Security Standards (DSS) that must be adhered to by systems handling payment information of any of their member brands (which include Visa, Mastercard, and several others). ISO is the International Standards Organization, and since they have so many completely different certifications, this does not tell you much. 

Reference: Cloud Compliance & Regulations Resources | Google Cloud


Question 43: Your new client has advised you that their organization deals with GDPR. What can you infer about their information systems?

A. Their system deals with medical information.

B. Their customers located in the EU may require them to delete their user data and provide evidence of such.

C. They will also need to pass a SOX audit.

D. They handle money-linked information.

ANSWER43:

B

Notes/References43:

SOX stands for Sarbanes Oxley and is US regulation governing financial reporting for publicly-traded companies. HIPAA–the Health Insurance Portability and Accountability Act of 1996–is US regulation aimed at safeguarding individuals' (i.e. patients’) health information. PCI is the Payment Card Industry, and they have Data Security Standards (DSS) that must be adhered to by systems handling payment information of any of their member brands (which include Visa, Mastercard, and several others). 

Reference: Cloud Compliance & Regulations Resources | Google Cloud

Question 44: For this question, refer to the Company C case study. Once Company C has completed their initial cloud migration as described in the case study, which option would represent the quickest way to migrate their production environment to GCP?

A. Apply the strangler pattern to their applications and reimplement one piece at a time in the cloud

B. Lift and shift all servers at one time

C. Lift and shift one application at a time

D. Lift and shift one server at a time

E. Set up cloud-based load balancing then divert traffic from the DC to the cloud system

F. Enact their disaster recovery plan and fail over

ANSWER44:

F

Notes/References44:

The proposed Lift and Shift options are all talking about different situations than Dress4Win would find themselves in, at that time: they’d then have automation to build a complete prod system in the cloud, but they’d just need to migrate to it. “Just”, right? 🙂 The strangler pattern approach is similarly problematic (in this case), in that it proposes a completely different cloud migration strategy than the one they’ve almost completed. Now, if we purely consider the kicker’s key word “quickest”, using the DR plan to fail over definitely seems like it wins. Setting up an additional load balancer and migrating slowly/carefully would take more time. 

Reference: Strangler pattern – Cloud Design Patterns | Microsoft Docs StranglerFigApplication Monolith to Microservices Using the Strangler Pattern – DZone Microservices Understanding Lift and Shift and If It’s Right For You

Question 45: Which of the following commands is most likely to appear in an environment setup script?

A. gsutil mb -l asia gs://${project_id}-logs

B. gcloud compute instances create –zone–machine-type=n1-highmem-16 newvm

C. gcloud compute instances create –zone–machine-type=f1-micro newvm

D. gcloud compute ssh ${instance_id}

E. gsutil cp -r gs://${project_id}-setup ./install

F. gsutil cp -r logs/* gs://${project_id}-logs/${instance_id}/

ANSWER45:

A

Notes/References45:

The context here indicates that “environment” is an infrastructure environment like “staging” or “prod”, not just a particular command shell. In that sort of a situation, it is likely that you might create some core per-environment buckets that will store different kinds of data like configuration, communication, logging, etc. You're not likely to be creating, deleting, or connecting (sshing) to instances, nor copying files to or from any instances. 

Reference: mb – Make buckets | Cloud Storage | Google Cloud cp – Copy files and objects | Cloud Storage | Google Cloud gcloud compute instances | Cloud SDK Documentation | Google Cloud

Question 46: Your developers are working to expose a RESTful API for your company’s physical dealer locations. Which of the following endpoints would you advise them to include in their design?

A. /dealerLocations/get

B. /dealerLocations

C. /dealerLocations/list

D. Source and destination

E. /getDealerLocations

ANSWER46:

B

Notes/References46:

It might not feel like it, but this is in scope and a fair question. Google expects Professional Cloud Architects to be able to advise on designing APIs according to best practices (check the exam guide!). In this case, it's important to know that RESTful interfaces (when properly designed) use nouns for the resources identified by a given endpoint. That, by itself, eliminates most of the listed options. In HTTP, verbs like GET, PUT, and POST are then used to interact with those endpoints to retrieve and act upon those resources. To choose between the two noun-named options, it helps to know that plural resources are generally already understood to be lists, so there should be no need to add another “/list” to the endpoint. 

Reference: RESTful API Design — Step By Step Guide – By


Question 47: Which of the following commands is most likely to appear in an instance shutdown script?

A. gsutil cp -r gs://${project_id}-setup ./install

B. gcloud compute instances create –zone–machine-type=n1-highmem-16 newvm

C. gcloud compute ssh ${instance_id}

D. gsutil mb -l asia gs://${project_id}-logs

E. gcloud compute instances delete ${instance_id}

F. gsutil cp -r logs/* gs://${project_id}-logs/${instance_id}/

G. gcloud compute instances create –zone–machine-type=f1-micro newvm

ANSWER47:

F

Notes/References47:

The startup and shutdown scripts run on an instance at the time when that instance is starting up or shutting down. Those situations do not generally call for any other instances to be created, deleted, or connected (sshed) to. Also, those would be a very unusual time to make a Cloud Storage bucket, since buckets are the overall and highly-scalable containers that would likely hold the data for all (or at least many) instances in a given project. That said, instance shutdown time may be a time when you'd want to copy some final logs from the instance into some project-wide bucket. (In general, though, you really want to be doing that kind of thing continuously and not just at shutdown time, in case the instance shuts down unexpectedly and not in an orderly fashion that runs your shutdown script.)

Reference:  Running startup scripts | Compute Engine Documentation | Google Cloud Running shutdown scripts | Compute Engine Documentation | Google Cloud cp – Copy files and objects | Cloud Storage | Google Cloud gcloud compute instances | Cloud SDK Documentation | Google Cloud

Question 48: It is Saturday morning and you have been alerted to a serious issue in production that is both reducing availability to 95% and corrupting some data. Your monitoring tools noticed the issue 5 minutes ago and it was just escalated to you because the on-call tech in line before you did not respond to the page. Your system has an RPO of 10 minutes and an RTO of 120 minutes, with an SLA of 90% uptime. What should you do first?

A. Escalate the decision to the business manager responsible for the SLA

B. Take the system offline

C. Revert the system to the state it was in on Friday morning

D. Investigate the cause of the issue

ANSWER48:

B

Notes/References48:

The data corruption is your primary concern, as your Recovery Point Objective allows only 10 minutes of data loss and you may already have lost 5. (The data corruption means that you may well need to roll back the data to before that started happening.) It might seem crazy, but you should as quickly as possible stop the system so that you do not lose any more data. It would almost certainly take more time than you have left in your RPO to properly investigate and address the issue, but you should then do that next, during the disaster response clock set by your Recovery Time Objective. Escalating the issue to a business manager doesn't make any sense. And neither does it make sense to knee-jerk revert the system to an earlier state unless you have some good indication that doing so will address the issue. Plus, we'd better assume that “revert the system” refers only to the deployment and not the data, because rolling the data back that far would definitely violate the RPO. 

Reference: Disaster recovery – Wikipedia

Question 49: Which of the following are not processes or practices that you would associate with DevOps?

A. Raven-test the candidate

B. Obfuscate the code

C. Only one of the other options is made up

D. Run the code in your cardinal environment

E. Do a canary deploy

ANSWER49:

A and D

Notes/References49:

Testing your understanding of development and operations in DevOps. In particular, you need to know that a canary deploy is a real thing and it can be very useful to identify problems with a new change you're making before it is fully rolled out to and therefore impacts everyone. You should also understand that “obfuscating” code is a real part of a release process that seeks to protect an organization's source code from theft (by making it unreadable by humans) and usually happens in combination with “minification” (which improves the speed of downloading and interpreting/running the code). On the other hand, “raven-testing” isn't a thing, and neither is a “cardinal environment”. Those bird references are just homages to canary deployments.

Reference: Intro to deployment strategies: blue-green, canary, and more – DEV Community ‍‍

Question 50: Your CTO is going into budget meetings with the board, next month, and has asked you to draw up plans to optimize your GCP-based systems for capex. Which of the following options will you prioritize in your proposal?

A. Object lifecycle management

B. BigQuery Slots

C. Committed use discounts

D. Sustained use discounts

E. Managed instance group autoscaling

F. Pub/Sub topic centralization

ANSWER50:

B and C

Notes/References50:

Pub/Sub usage is based on how much data you send through it, not any sort of “topic centralization” (which isn't really a thing). Sustained use discounts can reduce costs, but that's not really something you structure your system around. Now, most organizations prefer to turn Capital Expenditures into Operational Expenses, but since this question is instead asking you to prioritize CapEx, we need to consider the remaining options from the perspective of “spending” (or maybe reserving) defined amounts of money up-front for longer-term use. (Fair warning, though: You may still have some trouble classifying some cloud expenses as “capital” expenditures). With that in mind, GCE's Committed Use Discounts do fit: you “buy” (reserve/prepay) some instances ahead of time and then not have to pay (again) for them as you use them (or don't use them; you've already paid). BigQuery Slots are a similar flat-rate pricing model: you pre-purchase a certain amount of BigQuery processing capacity and your queries use that instead of the on-demand capacity. That means you won't pay more than you planned/purchased, but your queries may finish rather more slowly, too. Managed instance group autoscaling and object lifecycle management can help to reduce costs, but they are not really about capex. 

Reference: CapEx vs OpEx: Capital Expenses and Operating Expenses Explained – BMC Blogs Sustained use discounts | Compute Engine Documentation | Google Cloud Committed use discounts | Compute Engine Documentation | Google Cloud Slots | BigQuery | Google Cloud Autoscaling groups of instances | Compute Engine Documentation Object Lifecycle Management | Cloud Storage | Google Cloud


Question 51: In your last retrospective, there was significant disagreement voiced by the members of your team about what part of your system should be built next. Your scrum master is currently away, but how should you proceed when she returns, on Monday?

A. The scrum master is the one who decides

B. The lead architect should get the final say

C. The product owner should get the final say

D. You should put it to a vote of key stakeholders

E. You should put it to a vote of all stakeholders

ANSWER51:

C

Notes/References51:

In Scrum, it is the Product Owner's role to define and prioritize (i.e. set order for) the product backlog items that the dev team will work on. If you haven't ever read it, the Scrum Guide is not too long and quite valuable to have read at least once, for context. 

Reference: Scrum Guide | Scrum Guides

Question 52: Your development team needs to evaluate the behavior of a new version of your application for approximately two hours before committing to making it available to all users. Which of the following strategies will you suggest?

A. Split testing

B. Red-Black

C. A/B

D. Canary

E. Rolling

F. Blue-Green

G. Flex downtime

ANSWER52:

D and E

Notes/References52:

A Blue-Green deployment, also known as a Red-Black deployment, entails having two complete systems set up and cutting over from one of them to the other with the ability to cut back to the known-good old one if there’s any problem with the experimental new one. A canary deployment is where a new version of an app is deployed to only one (or a very small number) of the servers, to see whether it experiences or causes trouble before that version is rolled out to the rest of the servers. When the canary looks good, a Rolling deployment can be used to update the rest of the servers, in-place, one after another to keep the overall system running. “Flex downtime” is something I just made up, but it sounds bad, right? A/B testing–also known as Split testing–is not generally used for deployments but rather to evaluate two different application behaviours by showing both of them to different sets of users. Its purpose is to gather higher-level information about how users interact with the application. 

Reference: BlueGreenDeployment design patterns – What's the difference between Red/Black deployment and Blue/Green Deployment? – Stack Overflow design patterns – What's the difference between Red/Black deployment and Blue/Green Deployment? – Stack Overflow What is rolling deployment? – Definition from WhatIs.com A/B testing – Wikipedia

Question 53: You are mentoring a Junior Cloud Architect on software projects. Which of the following “words of wisdom” will you pass along?

A. Identifying and fixing one issue late in the product cycle could cost the same as handling a hundred such issues earlier on

B. Hiring and retaining 10X developers is critical to project success

C. A key goal of a proper post-mortem is to identify what processes need to be changed

D. Adding 100% is a safe buffer for estimates made by skilled estimators at the beginning of a project

E. A key goal of a proper post-mortem is to determine who needs additional training

ANSWER53:

A and C

Notes/References53:

There really can be 10X (and even larger!) differences in productivity between individual contributors, but projects do not only succeed or fail because of their contributions. Bugs are crazily more expensive to find and fix once a system has gone into production, compared to identifying and addressing that issue right up front–yes, even 100x. A post-mortem should not focus on blaming an individual but rather on understanding the many underlying causes that led to a particular event, with an eye toward how such classes of problems can be systematically prevented in the future. 

Reference: 403 Forbidden 403 Forbidden Google – Site Reliability Engineering The Cone of Uncertainty

Question 54: Your team runs a service with an SLA to achieve p99 latency of 200ms. This month, your service achieved p95 latency of 250ms. What will happen now?

A. The next month’s SLA will be increased.

B. The next month’s SLO will be reduced.

C. Your client(s) will have to pay you extra.

D. You will have to pay your client(s).

E. There is no impact on payments.

F. There is not enough information to make a determination.

ANSWER54:

D

Notes/References54:

It would be highly unusual for clients to have to pay extra, even if the service performs better than agreed by the SLA. SLAs generally set out penalties (i.e. you pay the client) for below-standard performance. While SLAs are external-facing, SLOs are internal-facing and do not generally relate to performance penalties. Neither SLAs nor SLOs are adaptively changed just because of one month’s performance; such changes would have to happen through rather different processes. A p99 metric is a tougher measure than p95, and p95 is tougher than p90–so meeting the tougher measure would surpass a required SLA, but meeting a weaker measure would not give enough information to say. 

Reference: What's the Difference Between DevOps and SRE? (class SRE implements DevOps) – YouTube Percentile rank – Wikipedia


Question 55: Your team runs a service with an SLO to achieve p90 latency of 200ms. This month, your service achieved p95 latency of 250ms. What will happen now?

A. The next month’s SLA will be increased.

B. There is no impact on payments.

C. There is not enough information to make a determination.

D. Your client(s) will have to pay you extra.

E. The next month’s SLO will be reduced.

F. You will have to pay your client(s).

ANSWER55:

B

Notes/References55:

It would be highly unusual for clients to have to pay extra, even if the service performs better than agreed by the SLA. SLAs generally set out penalties (i.e. you pay the client) for below-standard performance. While SLAs are external-facing, SLOs are internal-facing and do not generally relate to performance penalties. Neither SLAs nor SLOs are adaptively changed just because of one month’s performance; such changes would have to happen through rather different processes. A p99 metric is a tougher measure than p95, and p95 is tougher than p90–so meeting the tougher measure would surpass a required SLA, but meeting a weaker measure would not give enough information to say. 

Reference: What's the Difference Between DevOps and SRE? (class SRE implements DevOps) – YouTube Percentile rank – Wikipedia

Question 56: For this question, refer to the Company C case study. How would you recommend Company C address their capacity and utilization concerns?

A. Configure the autoscaling thresholds to follow changing load

B. Provision enough servers to handle trough load and offload to Cloud Functions for higher demand

C. Run cron jobs on their application servers to scale down at night and up in the morning

D. Use Cloud Load Balancing to balance the traffic highs and lows

D. Run automated jobs in Cloud Scheduler to scale down at night and up in the morning

E. Provision enough servers to handle peak load and sell back excess on-demand capacity to the marketplace

ANSWER56:

A

Notes/References56:

The case study notes, “Our traffic patterns are highest in the mornings and weekend evenings; during other times, 80% of our capacity is sitting idle.” Cloud Load Balancing could definitely scale itself to handle this type of load fluctuation, but it would not do anything to address the issue of having enough application server capacity. Provisioning servers to handle peak load is generally inefficient, but selling back excess on-demand capacity to the marketplace just isn’t a thing, so that option must be eliminated, too. Using Cloud Functions would require a different architectural approach for their application servers and it is generally not worth the extra work it would take to coordinate workloads across Cloud Functions and GCE–in practice, you’d just use one or the other. It is possible to manually effect scaling via automated jobs like in Cloud Scheduler or cron running somewhere (though cron running everywhere could create a coordination nightmare), but manual scaling based on predefined expected load levels is far from ideal, as capacity would only very crudely match demand. Rather, it is much better to configure the managed instance group’s autoscaling to follow demand curves–both expected and unexpected. A properly-architected system should rise to the occasion of unexpectedly going viral, and not fall over. 

Reference: Load Balancing | Google Cloud Google Cloud Platform Marketplace Solutions Cloud Functions | Google Cloud Cloud Scheduler | Google Cloud


Google Cloud Latest News, Questions and Answers online:

Cloud Run vs App Engine: In a nutshell, you give Google’s Cloud Run a Docker container containing a webserver. Google will run this container and create an HTTP endpoint. All the scaling is automatically done for you by Google. Cloud Run depends on the fact that your application should be stateless. This is because Google will spin up multiple instances of your app to scale it dynamically. If you want to host a traditional web application this means that you should divide it up into a stateless API and a frontend app.

With Google’s App Engine you tell Google how your app should be run. The App Engine will create and run a container from these instructions. Deploying with App Engine is super easy. You simply fill out an app.yml file and Google handles everything for you.

With Cloud Run, you have more control. You can go crazy and build a ridiculous custom Docker image, no problem! Cloud Run is made for Devops engineers, App Engine is made for developers. Read more here…


Cloud Run VS Cloud Functions: What to consider?

The best choice depends on what you want to optimize, your use-cases and your specific needs.

If your objective is the lowest latency, choose Cloud Run.

Indeed, Cloud Run use always 1 vCPU (at least 2.4Ghz) and you can choose the memory size from 128Mb to 2Gb.

With Cloud Functions, if you want the best processing performance (2.4Ghz of CPU), you have to pay 2Gb of memory. If your memory footprint is low, a Cloud Functions with 2Gb of memory is overkill and cost expensive for nothing.

Cutting cost is not always the best strategy for customer satisfaction, but business reality may require it. Anyway, it highly depends of your use-case

Both Cloud Run and Cloud Function round up to the nearest 100ms. As you could play with the GSheet, the Cloud Functions are cheaper when the processing time of 1 request is below the first 100ms. Indeed, you can slow the Cloud Functions vCPU, with has for consequence to increase the duration of the processing but while staying under 100ms if you tune it well. Thus less Ghz/s are used and thereby you pay less.

the cost comparison between Cloud Functions and Cloud Run goes further than simply comparing a pricing list. Moreover, on your projects, you often will have to use the 2 solutions for taking advantage of their strengths and capabilities.

My first choice for development is Cloud Run. Its portability, its testability, its openess on the libraries, the languages and the binaries confer it too much advantages for, at least, a similar pricing, and often with a real advantage in cost but also in performance, in particular for concurrent requests. Even if you need the same level of isolation of Cloud functions (1 instance per request), simply set the concurrent param to 1!

In addition, the GA of Cloud Run is applied on all containers, whatever the languages and the binaries used. Read more here…


What does the launch of Google’s App Maker mean for professional app developers?

Should I go with AWS Elastic Beanstalk or Google App Engine (Managed VMs) for deploying my Parse-Server backend?

Why can a company such as Google sell me a cloud gaming service where I can “rent” GPU power over miles of internet, but when I seek out information on how to create a version of this everyone says that it is not possible or has too much latency?

AWS wins hearts of developers while Azure those of C-levels. Google is a black horse with special expertise like K8s and ML. The cloud world is evolving. Who is the winner in the next 5 years?

What is GCP (Google Cloud Platform) and how does it work?

What is the maximum amount of storage that you could have in your Google drive?

How do I deploy Spring Boot application (Web MVC) on Google App Engine(GAE) or HEROKU using Eclipse IDE?

What are some downsides of building softwares on top of Google App Engine?

Why is Google losing the cloud computing race?

How did new products like Google Drive, Microsoft SkyDrive, Yandex.Disk and other cloud storage solutions affect Dropbox’s growth and revenues?

What is the capacity of Google servers?

What is the Hybrid Cloud platform?

What is the difference between Docker and Google App engines?

How do I get to cloud storage?

How does Google App Engine compare to Heroku?

What is equivalent of Google Cloud BigTable in Microsoft Azure?

How big is the storage capacity of Google organization and who comes second?

It seems strange that Google Cloud Platform offer “everything” except cloud search/inverted index?

Where are the files on Google Drive stored?

Is Google app engine similar to lambda?

Was Diane Greene a failure as the CEO of Google Cloud considering her replacement’s strategy and philosophy is the polar opposite?

How is Google Cloud for heavy real-time traffic? Is there any optimization needed for handling more than 100k RT?

When it comes to iCloud, why does Apple rely on Google Cloud instead of using their own data centers?


Google Cloud Storage : What bucket class for the best performance?: Multiregional buckets perform significantly better for cross-the-ocean fetches, however the details are a bit more nuanced than that. The performance is dominated by the latency of physical distance between the client and the cloud storage bucket.

  • If caching is on, and your access volume is high enough to take advantage of caching, there’s not a huge difference between the two offerings (that I can see with the tests). This shows off the power of Google’s Awesome CDN environment.
  • If caching is off, or the access volume is low enough that you can’t take advantage of caching, then the performance overhead is dominated directly by physics. You should be trying to get the assets as close to the clients as possible, while also considering cost, and the types of redundancy and consistency you’ll need for your data needs.

Top- high paying certifications:


  1. Google Certified Professional Cloud Architect – $139,529
  2. PMP® – Project Management Professional – $135,798
  3. Certified ScrumMaster® – $135,441
  4. AWS Certified Solutions Architect – Associate – $132,840
  5. AWS Certified Developer – Associate – $130,369
  6. Microsoft Certified Solutions Expert (MCSE): Server Infrastructure – $121,288
  7. ITIL® Foundation – $120,566
  8. CISM – Certified Information Security Manager – $118,412
  9. CRISC – Certified in Risk and Information Systems Control – $117,395
  10. CISSP – Certified Information Systems Security Professional – $116,900
  11. CEH – Certified Ethical Hacker – $116,306
  12. Citrix Certified Associate – Virtualization (CCA-V) – $113,442
  13. CompTIA Security+ – $110,321
  14. CompTIA Network+ – $107,143
  15. Cisco Certified Networking Professional (CCNP) Routing and Switching – $106,957

According to the 2020 Global Knowledge report, the top-paying cloud certifications for the year are (drumroll, please):

1- Google Certified Professional Cloud Architect — $175,761

2- AWS Certified Solutions Architect – Associate — $149,446

3- AWS Certified Cloud Practitioner — $131,465

4- Microsoft Certified: Azure Fundamentals — $126,653

5- Microsoft Certified: Azure Administrator Associate — $125,993

Sources:


1- Google Cloud

2- Linux Academy

3- WhizLabs

4- GCP Space on Quora

5- Udemy

6- Acloud Guru

7. Question and Answers are sent to us by good people all over the world.