We’re excited to announce the public preview of Automatic Zone Placement for Azure Virtual Machine Scale Sets. This new capability lets Azure select availability zones for your scale set based on SKU availability, capacity, and your placement requirements.
Automatic Zone Placement simplifies zonal deployments by removing the need to track SKU availability and maintain fixed zone lists for each Azure region. It is supported for scale sets using either Uniform or Flexible orchestration modes.
Simplify zonal deployments across regions
When creating a zonal scale set, you typically specify the availability zones where instances should be placed. However, SKU availability and capacity can vary between zones and regions. A configuration that works in one region might require a different zone selection in another, adding complexity to large-scale and multi-region deployments.
With Automatic Zone Placement, you define your placement requirements and allow Azure to select the zones that best satisfy them.
Key benefits
- Consistent multi-region configuration: Apply the same scale set configuration across regions without maintaining region-specific zone lists.
- Capacity-aware placement: Allow Azure to consider current capacity signals when selecting availability zones for deployment.
- Flexible resiliency controls: Configure how many zones the scale set can use and how much capacity can be concentrated in any one zone.
- Improved allocation flexibility: Allow the scale set to use additional eligible zones when its current zones cannot satisfy a scale-out request.
- Support for new and existing scale sets: Enable Automatic Zone Placement during scale set creation or migrate an existing scale set from customer-selected zones.
How Automatic Zone Placement works
When Automatic Zone Placement is enabled, Azure evaluates SKU availability, capacity, the scale set’s current zone distribution, and any placement constraints you configure before placing new VM instances.
If you enable Automatic Zone Placement without specifying additional placement settings, Azure uses the following default behavior:
- Targets an even distribution across three availability zones, with minimum spread across 2 zones.
- Limits any single zone to 50 percent of the scale set’s instances.
- Can expand into additional availability zones (3+) if the current zones cannot satisfy a scale-out request and another zone is available.
Any placement settings you provide take precedence where applicable. You can customize the eligible zones, maximum number of zones, maximum percentage of instances per zone, and balancing behavior to meet your workload requirements. These controls let you choose the appropriate tradeoff between allocation success and zonal distribution for your workload.
Automatic Zone Placement applies when Azure creates new instances. It does not move or recreate existing VMs. To proactively correct an imbalance among existing instances, you can use Automatic Zone Balance.
Get started
Automatic Zone Placement can be configured through the Azure portal, Azure CLI, Azure PowerShell, or the REST API.
To try the feature during public preview:
- Register the "Microsoft.Compute/VmssAutomaticZonePlacement" AFEC for your subscription.
- Use Compute API version 2026-03-01 or later.
- Create or update a scale set in a public Azure region that supports availability zones.
- Set the zone placement policy to Auto and configure any optional placement constraints.
For prerequisites, limitations, and configuration examples for each deployment method, see Configure automatic zone placement for a scale set.
Microsoft