AZ-900 Azure Compute Services
Azure compute is one of the most heavily tested domains on the
AZ-900 exam. Microsoft tests your ability to choose the right compute service for a given scenario — not deep technical configuration. This guide covers each major compute service, when to use it, and the IaaS vs PaaS distinction that underpins every exam question. Use it alongside our
AZ-900 practice tests to reinforce each concept before exam day.
Azure Virtual Machines
Azure Virtual Machines (VMs) are the foundational IaaS compute offering. When you create a VM, you get a virtualized server in the cloud where you control the operating system, installed software, security patches, and runtime environment. This maximum control comes with maximum responsibility — the
AZ-900 exam frequently tests this trade-off.
VMs are the right choice when you need to lift-and-shift an existing on-premises server to the cloud without refactoring the application, or when your workload requires a specific OS configuration that a managed service cannot provide.
Key VM concepts for AZ-900:
-
Availability Sets — protect VMs from rack-level failures by distributing across fault and update domains
-
Availability Zones — protect against datacenter failure by placing VMs in separate physical zones within a region
-
VM Scale Sets — automatically increase or decrease the number of VM instances based on demand
-
Azure Spot VMs — use unused Azure capacity at steep discounts, but can be evicted when capacity is needed
AZ-900 Study Tips
💡 What's the best study strategy for AZ-900?
Focus on weak areas first. Use practice tests to identify gaps, then study those topics intensively.
📅 How far in advance should I start studying?
Most successful candidates begin 4-8 weeks before the exam. Create a structured study schedule.
🔄 Should I retake practice tests?
Yes! Take each practice test 2-3 times. Focus on understanding why answers are correct, not memorizing.
✅ What should I do on exam day?
Arrive 30 min early, bring required ID, read questions carefully, flag difficult ones, and review before submitting.
App Service, Containers, and Kubernetes
When developers want to deploy web applications without managing servers, Azure App Service is the go-to PaaS solution. App Service handles OS patching, load balancing, auto-scaling, and SSL certificates automatically. You simply deploy your code (Node.js, Python, .NET, Java, PHP, or containers) and Azure runs it.
App Service is the AZ-900 answer whenever a scenario describes a company that wants to run a web app or REST API without worrying about the underlying infrastructure. It supports continuous deployment from GitHub, Azure DevOps, and other repositories.
For containerized workloads, Azure offers two core services:
Azure Container Instances (ACI) is the simplest way to run a container in Azure. There's no cluster to manage — you define the container image, CPU, and memory, and ACI starts it within seconds. ACI is PaaS and suits short-lived, single-container tasks, batch jobs, and testing scenarios.
Azure Kubernetes Service (AKS) is a fully managed Kubernetes orchestration platform. It suits complex, multi-container applications that need auto-scaling, rolling updates, service discovery, and high availability. AKS is the answer when an exam question mentions Kubernetes, microservices at scale, or orchestration.
Also covered in AZ-900 is Azure Virtual Desktop — a desktop and application virtualization service that lets users run a full Windows desktop environment from any device. It's often tested in scenarios involving remote work, bring-your-own-device (BYOD) policies, or centralizing application delivery. Read more about Azure deployment options in our AZ-900 complete guide.
Serverless Computing and Azure Functions
Serverless computing means you write and deploy code without thinking about servers at all — Azure allocates compute dynamically when your code runs and releases it immediately after. Azure Functions is the primary serverless offering on Azure.
Functions are triggered by events: an HTTP request, a message arriving in a storage queue, a timer schedule, a new file in blob storage, and many others. Each function execution is independent and stateless. Billing is based on the number of executions and their duration, so you pay nothing when your code is not running.
For AZ-900, the key serverless scenarios are:
- Running small pieces of code in response to events (HTTP trigger, timer, queue message)
- Building event-driven architectures without provisioning infrastructure
- Tasks where workload is unpredictable and you want automatic scaling to zero
Azure Functions can also be compared to Azure Logic Apps, which is a low-code/no-code workflow automation tool. Where Functions are code-first, Logic Apps are designer-first with hundreds of connectors to external services. Both are serverless, but exam questions distinguish them by whether a developer writes code or drags and drops connectors.
Understanding serverless is also key to the AZ-900 governance and compliance domain, since serverless services eliminate infrastructure you need to patch and audit.
Compute services connect directly to Azure's networking and storage layers. Virtual Machines attach to virtual networks and network security groups (covered in our AZ-900 networking guide). App Service and Functions store data in Azure Storage or databases (see AZ-900 storage services). Understanding how compute integrates with these layers is essential for scenario questions on the exam.
Test Your Compute KnowledgeAZ-900 Compute Exam Tips
AZ-900 Compute Checklist
Virtual Machines = IaaS: you manage OS, patches, and runtime — maximum control, maximum responsibility App Service = PaaS: deploy web apps and APIs without touching the OS or infrastructure Azure Functions = Serverless: event-triggered, pay per execution, scales to zero automatically ACI = simple single-container hosting, no cluster needed, fastest way to run a container AKS = managed Kubernetes for complex multi-container orchestration and auto-scaling Azure Virtual Desktop = cloud-hosted Windows desktop for remote users and BYOD scenarios VM Scale Sets enable automatic VM scaling based on metrics like CPU usage Availability Zones protect VMs from entire datacenter failures within a region AZ Pros and Cons
Pros
- AZ-900 has a defined, publicly available content blueprint — candidates know exactly what to prepare for
- Multiple preparation pathways (self-study, courses, coaching) accommodate different learning styles and schedules
- A growing ecosystem of study resources means candidates at any budget level can access quality preparation materials
- Clear score reporting allows candidates to identify specific strengths and weaknesses for targeted remediation
- Professional recognition associated with strong performance provides tangible career and academic benefits
Cons
- The scope of tested content requires substantial preparation time that competes with existing professional or academic commitments
- No single resource covers the full content scope — candidates typically need multiple study tools for comprehensive preparation
- Test anxiety and exam-day performance variability mean preparation effort does not always translate linearly to scores
- Registration, preparation, and potential retake costs accumulate into a significant financial investment
- Content and format can change between exam versions, making older preparation materials less reliable
AZ-900 Compute Questions and Answers
What is the difference between IaaS and PaaS compute on Azure?
IaaS (Infrastructure as a Service) gives you a virtual machine where you control the OS, patches, and installed software. Azure Virtual Machines are the main IaaS compute option. PaaS (Platform as a Service) lets you deploy your application without managing the underlying OS or infrastructure — Azure App Service and Azure Container Instances are PaaS examples. The AZ-900 exam tests this distinction frequently through scenario questions about responsibility and management overhead.
When should you use Azure Functions instead of a Virtual Machine?
Use Azure Functions when your workload is event-driven, short-running, and unpredictable in volume. Functions scale automatically from zero to millions of executions and bill only for actual compute time. Use a VM when you need continuous processing, require full OS control, or are migrating an existing server application that cannot be broken into event-driven pieces.
What is the difference between ACI and AKS?
Azure Container Instances (ACI) is a simple PaaS service for running individual containers quickly without managing any cluster infrastructure. Azure Kubernetes Service (AKS) is a fully managed Kubernetes platform for orchestrating multiple containers across a cluster with features like auto-scaling, rolling updates, and service discovery. ACI suits simple or short-lived container workloads; AKS suits complex, production-scale containerized applications.
Does Azure Virtual Desktop count as a compute service on AZ-900?
Yes. Azure Virtual Desktop is tested in the AZ-900 compute domain. It provides cloud-hosted Windows desktops and applications that users can access from any device. Exam questions typically frame it around remote work enablement, BYOD policies, or centralizing application management without distributing software to each endpoint.
What are VM Availability Zones and why do they matter?
Availability Zones are physically separate datacenters within an Azure region, each with independent power, cooling, and networking. Deploying VMs across multiple Availability Zones protects your application from a complete datacenter outage. This is distinct from Availability Sets, which protect against rack-level failures within a single datacenter. AZ-900 tests whether you know which option protects against which type of failure.
How does billing work for Azure Functions?
Azure Functions on the Consumption plan bill based on the number of function executions and the amount of memory used during each execution (measured in GB-seconds). The first 1 million executions per month are free. When your function is not running, you pay nothing. This pay-per-execution model is a defining characteristic of serverless computing and is contrasted with VMs, which bill continuously while running regardless of whether they are doing work.
Ready to test your knowledge? Our
free AZ-900 practice tests include dedicated compute scenario questions covering VM selection, PaaS vs IaaS trade-offs, serverless patterns, and container orchestration. Pair this guide with the AZ-900 complete guide for full exam coverage.