Azure Virtual Machine (VM)

An IaaS compute resource in Azure that provides a virtualized server (Windows or Linux) where you manage the operating system, runtime, and applications.

A logical grouping of VMs that distributes them across fault domains and update domains within a datacenter to reduce the impact of hardware failures and planned maintenance.

A physically separate location within an Azure region, with independent power, cooling, and networking. Deploying resources across zones increases resilience to datacenter-level failures.

Virtual Machine Scale Set (VMSS)

An Azure compute resource that lets you deploy and manage a set of identical, autoscaling VMs as a single resource, often used for stateless workloads.

A lightweight, portable unit that packages an application and its dependencies, sharing the host OS kernel and enabling consistent execution across environments.

A lightweight, portable unit that packages an application and its dependencies together and shares the host operating system kernel, enabling consistent execution across environments.

What is Azure Functions?

A serverless compute service in Azure that runs event-driven code without requiring you to provision or manage servers, scaling automatically and often billed per execution.

What is Availability set?

A logical grouping of VMs that distributes them across fault domains and update domains within a datacenter to reduce the impact of hardware failures and planned maintenance.

Azure Compute Building Blocks: VMs, Containers, and Serverless — Mastering Microsoft Azure Fundamentals (AZ-900): Complete Exam-Ready Course

Q: A company wants to run existing Docker-based microservices in Azure with rolling updates, service discovery, and autoscaling. They have Kubernetes experience and want to keep using it. Which Azure compute option is the best fit?

Azure Kubernetes Service (AKS). AKS is designed to run and orchestrate containerized applications using Kubernetes. It supports rolling updates, autoscaling, and service discovery. VMs alone lack orchestration; Functions are for event-driven serverless code, not full microservice platforms; App Service runs web apps but does not provide full Kubernetes-style microservice orchestration.

Q: You are designing a solution that should process messages from a queue. The load is highly variable, and the business wants to pay only when processing occurs, without managing servers. Which option best meets these requirements?

Azure Functions triggered by the queue. Azure Functions with a queue trigger are ideal for event-driven workloads with variable load and pay-per-execution pricing. VM scale sets and AKS still require you to manage infrastructure and pay for provisioned capacity; a single large VM is the least flexible and will be underutilized when the queue is empty.

Orienting Yourself: Azure Compute in the Big Picture

Where Compute Fits

Every Azure solution needs compute. Your main choices are virtual machines, containers, and serverless. Each fits differently with IaaS, PaaS, and serverless models.

Link to Service Models

VMs are classic IaaS: you get virtualized servers, storage, networking on demand. Container platforms and serverless functions sit closer to PaaS.

Regions and Resources

Compute resources run in Azure regions and Availability Zones. They are Azure resources inside resource groups and subscriptions, and their placement impacts resilience.

Exam Mindset

AZ-900 questions often describe a scenario and ask you to choose VMs, containers, or serverless. This module trains that pattern-recognition skill.

Azure Virtual Machines: Your IaaS Building Block

What Is an Azure VM?

An Azure VM is a virtual server in the cloud. You choose Windows or Linux, size, disks, and networking, and you manage what happens inside the OS.

What You Control

You install patches, configure firewalls, add runtime frameworks, and deploy apps. This flexibility is powerful but increases your management effort.

When VMs Fit Best

Use VMs for legacy apps, software that needs specific drivers, or when you must control the OS and security stack in detail.

Exam Signal Words

Phrases like "lift-and-shift", "full control of OS", or "custom security software" usually point toward choosing Azure Virtual Machines.

Scaling and Availability for VMs: Availability Sets, Zones, and Scale Sets

Why One VM Is Not Enough

A single VM is a single point of failure. Azure offers availability sets, Availability Zones, and VM scale sets to improve resilience and scale.

Availability Sets

Availability sets spread VMs across fault and update domains, reducing impact from host failures and planned maintenance in a single region.

Availability Zones

Availability Zones are separate datacenter locations in a region. Placing VMs in different zones protects against an entire zone outage.

VM Scale Sets

VM scale sets manage groups of identical VMs and support autoscaling based on metrics like CPU or request count.

Example: Choosing VM Availability Options

Scenario Overview

An online ordering system moves to Azure. Two web servers and a database must be highly available and handle traffic spikes.

Designing the Web Tier

Use a VM scale set for web servers, spread instances across multiple Availability Zones, and front them with a load balancer or Application Gateway.

Handling the Database

Use a managed PaaS database with zone redundancy, or, if self-managed, deploy database VMs with availability and zone awareness.

Link to Exam Clues

Host failure → availability set. Datacenter/zone failure → Availability Zones. Autoscaling VMs → VM scale set.

Containers and Azure Kubernetes Service (AKS): High-Level View

What Is a Container?

A container bundles an app and its dependencies into a portable unit. It shares the host OS kernel, so it is lighter than a VM.

Key Azure Container Services

Azure Container Registry stores images. Azure Container Instances runs containers directly. Azure Kubernetes Service orchestrates containers at scale.

AKS in a Nutshell

AKS is a managed Kubernetes service. Azure manages the control plane; you manage worker nodes and your containerized applications.

When Containers Fit

Think containers for microservices, portability, and when you need orchestration features like rolling updates and autoscaling.

Serverless with Azure Functions: Code Without Servers

What Is Serverless?

Serverless means you run code without managing servers. Azure still uses servers, but you do not see or administer them.

Azure Functions Basics

Azure Functions let you write small pieces of code that run when triggered by events like HTTP calls, queue messages, or timers.

Benefits and Use Cases

You get automatic scaling and pay-per-execution pricing. Use Functions for APIs, background jobs, and integration glue between services.

Exam Clues

Phrases like "event-driven", "pay only when code runs", or "no server management" usually point to Azure Functions.

Comparing VMs, Containers, and Serverless in a Real Scenario

Photo App Scenario

A startup builds a photo-sharing app with a web front-end, upload API, and background image processing. How should they host each part?

All on VMs

Separate Azure VMs host the front-end, API, and worker. This gives full control but requires OS management and more scaling work.

Containers on AKS

All components become containers in AKS. This suits microservices and portability, but needs Kubernetes skills and cluster operations.

Serverless and PaaS Mix

App Service hosts the front-end, Azure Functions host the API and background processing. This minimizes infrastructure work and scales automatically.

Thought Exercise: Match Workloads to Compute Options

Work through these short thought exercises. Answer in your head or jot down choices; then compare to the suggested reasoning.

Legacy ERP system

Requirements: Runs on Windows Server 2012 with a custom driver. Vendor does not support containers. Needs OS-level antivirus and monitoring agents.
Best fit: Azure Virtual Machines.
Why: Needs specific OS and drivers plus full OS control.

Event-driven data processing

Requirements: When a CSV file is uploaded to Blob Storage, parse it and write data into a database. Usage is spiky; some hours see no uploads.
Best fit: Azure Functions.
Why: Event-driven, pay-per-execution, no servers to manage.

Microservices-based e-commerce platform

Requirements: Dozens of small services, independent deployment, rolling updates, traffic spikes during sales. Team already uses Docker and Kubernetes on-prem.
Best fit: Azure Kubernetes Service (AKS).
Why: Container orchestration at scale, matches existing skills.

Internal line-of-business web app

Requirements: Simple .NET web app, predictable traffic, no need for OS-level tweaks. Team wants to minimize ops.
Best fit: Likely Azure App Service (PaaS web apps), which sits between VMs and serverless.

Reflect: In each case, what words pushed you toward IaaS, PaaS, containers, or serverless? This is the same reasoning you will use on AZ-900 scenario questions.

Quick Check: Compute Model Basics

Test your understanding of when to use VMs, containers, or serverless.

A company wants to run existing Docker-based microservices in Azure with rolling updates, service discovery, and autoscaling. They have Kubernetes experience and want to keep using it. Which Azure compute option is the best fit?

Azure Virtual Machines in an availability set
Azure Kubernetes Service (AKS)
Azure Functions in the Consumption plan
Azure App Service (Web Apps)

Show Answer

Answer: B) Azure Kubernetes Service (AKS)

AKS is designed to run and orchestrate containerized applications using Kubernetes. It supports rolling updates, autoscaling, and service discovery. VMs alone lack orchestration; Functions are for event-driven serverless code, not full microservice platforms; App Service runs web apps but does not provide full Kubernetes-style microservice orchestration.

Quick Check: Availability and Serverless

Another short question to reinforce availability and serverless concepts.

You are designing a solution that should process messages from a queue. The load is highly variable, and the business wants to pay only when processing occurs, without managing servers. Which option best meets these requirements?

Azure Virtual Machines in a Virtual Machine Scale Set
Azure Kubernetes Service running a containerized worker
Azure Functions triggered by the queue
A single large Azure Virtual Machine