Microsoft Azure Compute
Microsoft Azure Compute is the set of Azure services that run virtual machines, web applications, containers, functions, and batch jobs. You choose among them by matching the workload to the control, scaling, and operational responsibility you need.
itCloud computing | OpenSkills.info
Course pathWalk it in order
Look it upDip in anytime
Go furtherLeaves this page
Don't Panic
Don't Panic: Microsoft Azure Compute
Azure Compute is the collection of Azure hosting models that run application code. The mildly inconvenient truth is that this does not name one machine-shaped thing. It names a choice between a virtual server, a managed web application, a container, a triggered function, or a pool of parallel work. Calling all of them "servers" is how an innocent planning meeting acquires extra operational work.
Start with the execution unit, meaning the thing Azure starts and scales. A Virtual Machine runs a Windows or Linux guest operating system. App Service runs a web application, API, mobile back end, or web job. Functions runs one invocation after a trigger. Container Apps runs a container app revision or job. AKS runs a Kubernetes workload. Container Instances runs a container group. Batch runs tasks from a queued job on a compute pool. The label matters because a function invocation and a VM do not fail, scale, or charge in the same way.
The useful map is the responsibility ladder. Virtual Machines provide host control, which includes configuring and patching the guest, protecting access, installing software, and planning capacity. App Service and Container Apps remove host work inside their application models. Functions narrows the arrangement to triggered code. AKS supplies Kubernetes API and control-plane access, then hands the cluster configuration, node pools, upgrades, workload policies, networking, observability, and application security back across the desk. It is a very capable desk.
The surprise is that managed hosting does not make the application managed. Durable state still needs a service built to hold it. Health checks still need to describe readiness. Identities, permissions, network exposure, secrets, recovery, deployment, and rollback still require deliberate choices. A single VM is one instance, not a resilience strategy wearing a cloud badge. More instances only help when they can reduce the constraint that is actually causing pressure.
Read the intro when you need the full selection and reliability model. Use the slides to compare execution units and control boundaries at a glance. Keep the cheatsheet nearby when choosing placement, scaling signals, security ownership, and pre-deployment checks. The reference tab is the route into the Azure service documentation once the workload has earned a specific candidate.
Where this skill leads
Relevant careers
See how this topic contributes to broader role-level skill maps.
Sources
- https://learn.microsoft.com/en-us/azure/architecture/guide/technology-choices/compute-decision-tree
Supports
- Azure compute is the hosting model for resources that run application code
- Virtual Machines, App Service, Functions, AKS, Container Apps, Container Instances, Batch, and Azure VMware Solution serve different workload models
- Infrastructure, platform, and functions as a service trade control against management responsibility
- Container Apps hides the Kubernetes API while AKS provides Kubernetes API and control plane access
- General-purpose Azure compute platforms are regional and require external routing across regional deployments
- https://learn.microsoft.com/en-us/azure/virtual-machines/overview
Supports
- Azure VMs provide Windows or Linux virtualized compute with guest operating system control
- VM operators configure, patch, and install guest software
- VM size determines processor, memory, storage, and network capacity
- VMs use related networking and storage resources
- Availability zones and scale sets provide availability and scaling options
- https://learn.microsoft.com/en-us/azure/virtual-machine-scale-sets/overview
Supports
- Virtual Machine Scale Sets create and manage groups of load-balanced VMs
- Scale sets can increase or decrease instance count in response to demand or schedules
- Scale sets support fault-domain and availability-zone placement
- Scale sets centralize configuration, updates, and instance management
- https://learn.microsoft.com/en-us/azure/virtual-machines/availability-set-overview
Supports
- Availability sets distribute VMs across fault and update domains
- Availability zones provide greater isolation than availability sets
- Availability sets do not protect against guest operating system or application failures
- https://learn.microsoft.com/en-us/azure/app-service/overview
Supports
- App Service hosts web applications, mobile back ends, and REST APIs
- App Service supports managed language runtimes and custom containers
- App Service provides deployment, scaling, networking, and platform management features
- https://learn.microsoft.com/en-us/azure/azure-functions/functions-overview
Supports
- Azure Functions is an event-driven serverless compute service
- Triggers start functions and bindings connect functions to other services
- Functions supports HTTP, schedule, queue, file, stream, and database-change scenarios
- Functions hosting plans provide different scaling, networking, warm-instance, duration, and billing characteristics
- https://learn.microsoft.com/en-us/azure/container-apps/overview
Supports
- Container Apps is a managed serverless platform for containerized applications
- Container Apps provides revisions, replicas, ingress, scaling, and jobs
- Container Apps abstracts Kubernetes infrastructure and API operations
- https://learn.microsoft.com/en-us/azure/aks/what-is-aks
Supports
- AKS is a managed Kubernetes service
- Azure manages the Kubernetes control plane
- Customers manage workload configuration, node pools, networking, upgrades, security, and observability responsibilities
- https://learn.microsoft.com/en-us/azure/container-instances/container-instances-overview
Supports
- Container Instances runs Linux or Windows containers without VM provisioning
- Container Instances does not require adoption of a higher-level orchestrator
- A container group is the scheduling and resource unit
- https://learn.microsoft.com/en-us/azure/batch/batch-technical-overview
Supports
- Azure Batch runs large-scale parallel and high-performance computing applications
- Batch uses pools of compute nodes, jobs, and tasks
- Batch schedules tasks and can scale compute pools
- https://learn.microsoft.com/en-us/azure/architecture/best-practices/auto-scaling
Supports
- Autoscaling uses workload metrics and rules to adjust resource capacity
- Scaling policies require capacity bounds, cooldown behavior, and testing
- Scaling cannot resolve a bottleneck that added compute capacity does not affect
- https://learn.microsoft.com/en-us/azure/security/fundamentals/shared-responsibility
Supports
- Microsoft operates physical infrastructure and managed service layers
- Customer responsibility changes across infrastructure, platform, and software service models
- Customers retain responsibility for data, identities, accounts, and access management
- https://learn.microsoft.com/en-us/entra/identity/managed-identities-azure-resources/overview
Supports
- Managed identities let Azure resources authenticate without application-managed credentials
- Managed identities are identities in Microsoft Entra ID
- https://learn.microsoft.com/en-us/azure/azure-monitor/fundamentals/overview
Supports
- Azure Monitor collects and analyzes metrics, logs, traces, and events
- Azure Monitor supports alerts and actions based on telemetry
- https://azure.microsoft.com/en-us/blog/azure-app-service-improves-node-js-and-php-developer-experience/
Supports
- Azure App Service entered general availability in March 2015.
- https://azure.microsoft.com/en-us/blog/azure-vm-scale-sets-public-preview/
Supports
- Virtual Machine Scale Sets entered public preview in November 2015 and became generally available in April 2016.
- https://azure.microsoft.com/en-us/blog/announcing-general-availability-of-azure-functions//
Supports
- Azure Functions became generally available in November 2016 after a March 2016 preview.
- https://azure.microsoft.com/en-us/blog/announcing-general-availability-of-managed-disks-and-larger-scale-sets/
Supports
- Azure announced general availability of Managed Disks and larger Virtual Machine Scale Sets in February 2017.
- https://azure.microsoft.com/en-us/blog/general-availability-of-app-service-on-linux-and-web-app-for-containers/
Supports
- Azure App Service on Linux and Web App for Containers became generally available in September 2017.
- https://azure.microsoft.com/en-us/blog/azure-container-instances-now-generally-available/
Supports
- Azure Container Instances became generally available in April 2018.
- https://azure.microsoft.com/en-us/blog/azure-kubernetes-service-aks-ga-new-regions-more-features-increased-productivity/
Supports
- Azure Kubernetes Service became generally available in June 2018.
- https://azure.microsoft.com/en-us/blog/azure-virtual-machine-scale-sets-now-provide-simpler-management-during-scalein/
Supports
- Azure announced general availability of VM Scale Set instance protection, custom scale-in policy, and terminate notification in May 2020.
- https://azure.microsoft.com/en-us/updates?id=generally-available-azure-container-apps
Supports
- Azure Container Apps became generally available in May 2022.
- https://blog.aks.azure.com/2026/07/28/nap-disruption
Supports
- AKS node auto-provisioning can remove nodes during consolidation or drift replacement; workload disruption budgets and maintenance windows regulate that work.
- https://techcommunity.microsoft.com/blog/microsoftmissioncriticalblog/your-on-call-rotation-has-a-new-member-10-production-incidents-end-to-end-with-a/4545187
Supports
- A VM Scale Set configuration extension failure can leave an instance running while its application health probes fail, reducing healthy backend capacity and causing intermittent errors.