Microsoft announced the public preview of automatic zone balancing for Azure Virtual Machine Scale Sets, which expands the resiliency toolkit for enterprise cloud deployments. This feature is built to proactively protect your workloads by maintaining a balanced VM distribution and minimizing the impact of unexpected availability zone failures.

Azure Virtual Machine Scale Set (VMSS) Provides a way to deploy and manage large groups of identical virtual machines. The service automatically distributes them across your infrastructure to improve availability, simplify management, and support dynamic scaling based on demand. This allows administrators to centrally configure, update, and load balance thousands of VMs while ensuring high uptime through features such as availability zones, autoscaling, and integrated orchestration tools.

How does automatic rebalance protect against zone failure?

The goal of Azure VM scale sets is to virtual machine Evenly distributed across availability zones for increased resiliency. However, real-world factors such as expanded activity and capacity constraints can lead to an uneven distribution over time, increasing the risk that a failure in one zone will impact an organization’s workloads more than intended. Automatic zone balance addresses this issue by continuously checking for these changes and silently redistributing VMs in the background to restore a uniform and resilient layout.

“Automatic zone balancing addresses these challenges by continuously monitoring the scale set’s zone imbalances and tracking capacity in under-provisioned zones. VMs are automatically created in under-provisioned zones when capacity becomes available, eliminating the need for manual trial and error.” explained.

How does automatic zone balance work?

Automatic zone balancing helps distribute virtual machine scale sets evenly across two or more availability zones by continuously monitoring and remediating uneven placement of VM instances with minimal impact. When it detects that one zone has fewer VMs than other zones, it restores balance by adding new instances to the zone with the lowest number and checking health before making changes to other zones.

Once the newly created VM is verified to be healthy, the system removes the instance from the zone with the highest number of VMs. This shifts only one VM at a time, keeping the running workload stable throughout the process. This method uses a create-before-delete workflow to always provision new VMs with the latest SKU defined in the scale set model. This allows you to maintain consistency and resilience while rebalancing your environment.

How do I get started?

Microsoft said that enabling automatic zone balancing also automatically turns on automatic instance repair, which combines zone-level resiliency with instance-level health monitoring. This reduces operational overhead, reduces risk during zone failures, and improves workload resiliency.

To participate in the public preview, admins can automatic zone rebalance Set feature flags in your subscription via the Azure portal, CLI, or PowerShell. A scale set must first meet several requirements. That is, it must run across at least two availability zones, be configured to use best-effort zone distribution mode, and have application health monitoring enabled.