Interview Questions
Azure Interview Questions and Answers
Azure interviews often lean on its strong integration with the Microsoft/.NET ecosystem and enterprise identity (Azure AD) -- worth being comfortable with even if your background is primarily AWS.
Example: An Azure Resource Manager (ARM/Bicep) snippet
Bicepresource appService 'Microsoft.Web/sites@2022-03-01' = {
name: 'my-api-app'
location: resourceGroup().location
properties: {
serverFarmId: appServicePlan.id
httpsOnly: true
}
}
// Infrastructure-as-code: this App Service definition is version-controlled
// alongside application code, rather than clicked together manually in the portal.Frequently Asked Questions
A fully managed platform for hosting web apps and APIs -- you deploy your code (or a container), and Azure handles the underlying servers, OS patching, and scaling. It's Azure's rough equivalent to a PaaS layer above raw virtual machines, similar in spirit to AWS Elastic Beanstalk.
A Subscription is the billing and access-control boundary -- everything in it is billed together and can have subscription-wide policies. A Resource Group is a logical container within a subscription for grouping related resources (e.g. all resources for one application) so they can be managed, deployed, and deleted together.
Microsoft's cloud identity and access management service -- it handles user authentication and authorization, and is deeply integrated across Azure services and Microsoft 365. It's commonly used both for securing access to Azure resources themselves and as an identity provider for an organization's own applications (via OAuth2/OIDC).
Azure Functions is serverless -- you deploy individual functions that run in response to triggers (HTTP request, timer, queue message) and pay only for execution time. App Service runs a continuously deployed application (a whole API or web app), always running, with more control over the runtime environment.
It's Azure's private network boundary for your resources -- similar to an AWS VPC. It lets you control IP address ranges, subnet resources logically, and restrict which resources can communicate with each other or with the internet, rather than having everything exposed by default.
Azure SQL Database is a fully managed PaaS database service -- Microsoft handles patching, backups, and high availability. SQL Server on a VM is IaaS -- you install and manage SQL Server yourself on a virtual machine you control, trading more management overhead for more configuration flexibility (useful when you need a specific SQL Server feature or version not available in the managed offering).
Pay-as-you-go by default (billed per second/minute of actual usage), with Reserved Instances (commit to 1 or 3 years for a significant discount) and Spot pricing (heavily discounted, but can be reclaimed by Azure with short notice, suited to interruption-tolerant workloads) as cost-optimization options for predictable or flexible workloads respectively.