Cloud Computing Midterm Exam

Reviewer

Deploy a Hybrid Cloud deployment keeping primary medical records in the private infrastructure while using secure hybrid interconnects to feed anonymized data into a public cloud ML engine.

Scenario: A legacy healthcare enterprise running high-throughput transactional database systems on-premises must comply with strict national data residency mandates requiring medical records to never leave physical government-certified data centers. However, their data analytics team desperately requires elasticity and cloud-native machine learning services to train predictive models on anonymized data.

Which architecture represents the most appropriate, compliant cloud deployment strategy?

Platform as a Service (PaaS)

Scenario: A fintech startup is building a microservices-based mobile payment platform. They want their software engineers to focus exclusively on business logic, API integrations, and code optimization without spending any overhead on operating system patching, runtime environment maintenance, container orchestrator installation, or middleware updates.

Which cloud service model level should form the foundation of their development strategy?

The Customer, because OS-level application software maintenance, patching, and configuration fall under the customer's responsibility in IaaS.

Scenario: An enterprise running an IaaS Linux Virtual Machine on a public cloud provider experiences a major ransomware breach due to an unpatched vulnerability in an outdated Apache HTTP Web Server installed by their sysadmin team.

According to the Shared Responsibility Model, who bears primary responsibility for this security incident?

Rapid Elasticity and shift from CapEx (Capital Expenditure) to OpEx (Operational Expenditure)

Scenario: A global retail firm experiences massive server traffic spikes during Black Friday (1000x normal capacity), but experiences standard low-traffic baseline for the remaining 11 months of the year. Historically, maintaining physical data centers forced them to purchase hardware for peak capacity that sat idle most of the year.

Which core cloud characteristics and financial dynamics best resolve this inefficiency?

Availability Zones within a Cloud Region

Scenario: An e-commerce platform requires zero downtime. If an entire data center facility loses power or network access due to a localized disaster, the platform's database and application servers must fail over instantly to another physical facility within the same geographic region without data loss.

Which cloud architectural construct directly addresses this fault-tolerance requirement?

Management Groups

Scenario: A multinational corporation needs to structure its Azure environment. They want to ensure that compliance policies (such as restricting resource deployment to European regions) are enforced automatically across multiple Azure Subscriptions assigned to different business units (Finance, HR, Engineering).

What Azure management construct should be placed at the top of the hierarchy to enforce policy inheritance across all subscriptions?

Resource Group

Scenario: A DevOps team creates a cloud application environment in Azure consisting of 2 App Services, 1 SQL Database, and 1 Virtual Network. They want to treat these components as a single lifecycle unit so that when the testing phase ends, all associated resources can be deleted together in a single operation.

Where should these resources be logically grouped?

Azure Policy (Deny policy on VM SKUs)

Scenario: An IT manager discovers that junior developers have mistakenly provisioned massive, highly expensive GPU-enabled virtual machines in test environments, incurring thousands of dollars in unauthorized costs.

Which Azure tool should be implemented to prevent users from provisioning specific VM SKUs, regardless of their role permissions?

Azure ExpressRoute

Scenario: A enterprise requires a dedicated, private connection between its on-premises data center and Azure. The traffic MUST NOT traverse the public internet due to strict bandwidth guarantees and privacy compliance.

Which Azure networking service fulfills this requirement?

Azure Resource Manager (ARM) Templates / Bicep code

Scenario: A Lead Cloud Engineer wants to ensure that infrastructure provisioning is repeatable, version-controlled via Git, declarative, and idempotent across development, staging, and production environments in Azure.

Which methodology/tool aligns with best practices for this requirement?

Deploy two or more VMs across an Availability Set or across Availability Zones.

Scenario: A company is migrating a legacy database application to Azure VMs. The application requires direct access to OS registry settings, specific custom drivers, and predictable underlying CPU performance. However, during hardware maintenance events in Azure's data center, the company wants to ensure high availability with an SLA of 99.95%.

Azure App Service (Web Apps)

Scenario: A software development firm is launching a customer-facing Node.js web application. They expect traffic to fluctuate wildly throughout the day. They do not want to manage operating systems, web server software, or scale infrastructure manually.

Which Azure compute service offers auto-scaling, built-in deployment slots, SSL certificate management, and zero OS administration?

Azure Functions (Serverless)

Scenario: An image-processing system needs to run a small piece of code whenever a user uploads a photo to an Azure Blob Storage container. The code resizes the image and saves metadata to a database. The code takes 3 seconds to run, and the volume of uploads varies from zero at night to thousands per minute during peak hours. The company wants to pay ONLY for the execution time of the code.

Which Azure compute service is best suited for this event-driven workflow?

Azure Kubernetes Service (AKS)

Scenario: An enterprise is re-architecting its monolithic application into 50+ microservices packaged as Docker containers. They need automated service discovery, rolling updates, self-healing, horizontal pod autoscaling, and complex cluster management.

Azure Container Instances (ACI)

Scenario: A data analytics team needs to run an isolated, containerized batch job every midnight that runs for 15 minutes and then terminates. They do not want to provision a full Kubernetes cluster or manage underlying VM infrastructure for a job that runs so briefly.

What is the most cost-effective and low-friction Azure compute service for this task?

Azure Virtual Machine Scale Sets (VMSS) & GCP Compute Engine Managed Instance Groups (MIGs)

Scenario: A multi-cloud enterprise cloud team is translating an existing AWS infrastructure architecture to Azure and GCP. The AWS architecture heavily relies on AWS EC2 instances with Auto Scaling Groups for elastic VM scaling.

Which services across Azure and GCP directly correspond to AWS EC2 + Auto Scaling Groups?

AWS Lambda <-> Azure Functions <-> GCP Cloud Functions

Scenario: An architect is designing an event-driven serverless architecture that triggers code in response to message queue events. The company wants to compare serverless computing options across AWS, Azure, and GCP.

Which triplet correctly maps equivalent serverless compute services across all three providers?

AWS EKS, Azure AKS, GCP GKE

Scenario: A organization relies on Kubernetes for container orchestration and is planning a multi-cloud deployment across AWS, Azure, and Google Cloud Platform to prevent vendor lock-in.

What are the exact managed Kubernetes offerings provided by AWS, Azure, and GCP respectively?

AWS Spot Instances, Azure Spot Virtual Machines, GCP Preemptible VMs (Spot VMs)

Scenario: A financial company needs short-term, fault-tolerant batch compute capacity to perform overnight risk analysis. The workload can tolerate sudden machine terminations if the cloud provider reclaims capacity. Cost minimization is the absolute top priority.

Which compute pricing options should be utilized across AWS, Azure, and GCP?

Google Cloud Platform (GCP)

Scenario: A machine learning engineering firm is comparing GCP, AWS, and Azure for training ultra-large deep learning transformer models. They need specialized hardware designed natively for vector and matrix mathematics with custom TPU (Tensor Processing Unit) acceleration.

Which cloud provider natively offers custom TPU hardware specifically built for high-performance ML workloads?

3-Year Reserved Virtual Machine Instances (RI)

Scenario: A company operates 50 production database VMs in Azure that run continuously 24/7/365. The workload is highly predictable and will remain unchanged for at least 3 years.

Which pricing mechanism yields the maximum cost savings for this scenario?

Principle of Least Privilege using Azure RBAC with custom/built-in storage roles and Just-In-Time/PIM access.

Scenario: A enterprise contractor needs access to manage a specific Azure Storage Account for 2 weeks. The security administrator wants to ensure the contractor has ONLY the exact minimum permissions necessary to manage storage objects and that access automatically expires.

Which cloud security governance principle and Azure implementation should be applied?

Azure Arc

Scenario: A manufacturing company operates on-premises physical servers in remote factories with limited connectivity. They want to manage, monitor, and enforce Azure governance policies (such as Azure Policy and Defender for Cloud) across both their Azure cloud VMs and these on-premises physical servers from a unified pane of glass.

Which Azure technology enables managing on-premises infrastructure via the Azure control plane?

From On-Premises (Private) to Software as a Service (SaaS)

Scenario: An enterprise wants to migrate from an on-premises Microsoft Exchange email server to Microsoft 365 (Office 365).

This migration represents a shift between which two service model paradigms?

GCP enters user traffic into Google's private global fiber network backbone at the nearest edge point of presence (PoP) as early as possible.

Scenario: A cloud architect is designing a multi-cloud network routing strategy. They need to understand how network traffic flows between regions for Azure, AWS, and GCP.

What is a key architectural characteristic of Google Cloud Platform's global network compared to default public internet routing?