AZ-900 Scenario Guide: Choosing the Right Compute

To choose the right Azure compute service for AZ-900 scenarios, match the workload to the service: use Virtual Machines for legacy apps requiring full OS control, App Service for scalable web apps, Azure Functions for event-driven serverless tasks, and Azure Kubernetes Service (AKS) for complex container orchestration and management.

Why does the AZ-900 test your compute knowledge through scenarios?

If you've started studying for the AZ-900, you've probably noticed that Microsoft doesn't just ask you to define a service; they ask you to apply it. The exam objectives for 'Describe Azure Compute Services' are designed to see if you can distinguish between IaaS, PaaS, and Serverless in a real-world business context. It's one thing to know that a Virtual Machine is 'Infrastructure as a Service,' but it's another to know why a company would choose it over a container.

In my experience mentoring students, this is where most people stumble. They memorize the glossary but fail the scenario questions. To pass, you need to look for 'trigger words' in the question—words like 'full control,' 'event-driven,' or 'orchestration.' These are the clues that point you toward the correct answer. We focus heavily on these patterns at Cert Sensei to ensure you aren't just guessing based on a definition.

When should you choose Virtual Machines for legacy apps?

When you see a scenario mentioning 'legacy applications,' 'specific OS requirements,' or 'full administrative control,' your mind should immediately go to Azure Virtual Machines (VMs). Because VMs are IaaS, you are responsible for the OS, the patching, and the software configuration. This makes them the only viable choice for old software that requires a specific version of Windows Server or a custom Linux kernel that isn't supported by PaaS offerings.

For example, if a company needs to migrate a 10-year-old accounting app that requires a specific registry key change on the server, a VM is the answer. You get the 'virtual hardware,' and you handle the rest. Just remember: with great power comes great responsibility. In a VM scenario, the customer manages the OS; in almost every other compute scenario on the AZ-900, Microsoft handles the underlying platform for you.

How do you handle rapidly scaling web apps with App Service?

Azure App Service is the gold standard for PaaS (Platform as a Service) scenarios. If the scenario describes a developer who wants to 'deploy a web app quickly' without worrying about the underlying server, or a business that needs 'automatic scaling' to handle spikes in web traffic, App Service is your best bet. It abstracts the infrastructure, allowing you to focus entirely on the code.

Think of App Service as the 'middle ground.' You don't have the total control of a VM, but you have far more configuration options than a serverless function. When you see mentions of 'web apps,' 'REST APIs,' or 'mobile back-ends' that need to scale up or down based on demand, App Service is usually the correct choice. It removes the overhead of patching and OS management, which is a key selling point in many exam questions.

Which service fits event-driven microservices?

When a scenario mentions 'serverless,' 'event-driven,' or 'running a small piece of code in response to a trigger,' you are looking at Azure Functions. This is the peak of abstraction. You aren't managing a server, a platform, or even a container—you're just uploading a function. Common triggers include a file being uploaded to Blob Storage or a timer going off every hour.

One of the biggest hints for Azure Functions is the cost model. Because it's serverless, you only pay for the time the code is actually running. If a scenario emphasizes 'minimizing costs for a task that runs infrequently,' Azure Functions is the winner. It's perfect for microservices that don't need to be 'always on' but must react instantly to specific system events.

When is Azure Kubernetes Service (AKS) the right move?

Containers are a huge part of the AZ-900, and the distinction between simple containers and orchestrated containers is critical. If the scenario mentions 'managing a large number of containers,' 'microservices architecture,' or 'orchestration,' you need Azure Kubernetes Service (AKS). AKS is designed for complex environments where you need to manage how containers communicate, scale, and recover from failure.

Contrast this with Azure Container Instances (ACI). If the scenario asks for the 'simplest' way to run a single container without managing a cluster, choose ACI. But the moment the word 'orchestration' or 'cluster' appears, shift your answer to AKS. It's the heavy lifter of the Azure compute world, providing the automation needed to run enterprise-grade containerized applications at scale.

How can practice exams bridge the gap between theory and scenarios?

Reading the documentation is a start, but you won't truly master these scenarios until you've seen them from five different angles. This is why we built Cert Sensei. We provide 1,000 expert-curated Microsoft Azure Fundamentals (AZ-900) practice questions that mirror the actual exam's complexity. Instead of simple 'What is X?' questions, we give you the scenarios described in this guide, forcing you to apply your knowledge.

Beyond just the questions, our detailed expert reasoning explains *why* an answer is correct and why the other three are wrong—which is where the real learning happens. Plus, with our domain-level analytics, you can see if you're consistently missing compute questions. If your 'Compute' score is lagging while your 'Core Azure Services' score is high, you know exactly where to spend your next two hours of study time to ensure a first-time pass.

❓ Frequently Asked Questions