IaaS (managed VMware) · Not container-based

Azure VMware Solution

Run VMware workloads natively on Azure with a managed VMware private cloud.

When to use it

Choose Azure VMware Solution to run existing VMware (vSphere, vSAN, NSX) workloads natively on Azure with minimal change. It suits datacenter exit, capacity expansion, and disaster recovery where you want to keep VMware tooling and operations while gaining Azure proximity and scale.

When not to use it

Avoid Azure VMware Solution if you're building cloud-native or can re-platform to Azure-native services — those are more cost-effective and elastic. It's specifically for preserving an existing VMware investment, not for greenfield workloads.

Trade-offs at a glance

Hosting model IaaS (managed VMware)
Container support Not container-based
Minimum nodes 3
State management Stateless or stateful
Web hosting Agnostic
Autoscaling Built-in service
Load balancer Integrated (VMware NSX)
Scale limit 3 to 16 VMware ESXi hosts per vCenter
Multiregion Single region only — external router plus multiple instances
Virtual network integration Supported
Hybrid connectivity Supported
GPU support Not supported
TLS Set up in the VM
Architecture styles VM workloads based on the VMware format
Required skills VMware administration
Operational overhead Medium — VMware-managed infrastructure
Best for teams Teams with VMware platform requirements

Frequently asked questions

Does Azure VMware Solution run native VMware?

Yes. It runs the full VMware stack — vSphere, vSAN, and NSX — on dedicated Azure bare-metal hosts, so existing workloads and tooling carry over with little change.

What is the minimum host count?

A private cloud starts with three ESXi hosts and scales up; three is the minimum for a supported configuration.

When should I migrate off Azure VMware Solution to native Azure?

Once a workload can be re-platformed to App Service, AKS, or Azure VMs cost-effectively. AVS is ideal as a fast migration landing zone, with native modernization as a later step.

Starting point, not a verdict

This is one candidate from the Azure compute decision guide. The right choice depends on your full requirements — evaluate scaling, cost, and operational fit before committing.

Browse all tools →