Edit

Share via

Create a Virtual Machine Scale Set that uses Availability Zones

  • Article

Azure availability zones are fault-isolated locations within an Azure region that provide redundant power, cooling, and networking. They allow you to run applications with high availability and fault tolerance to data center failures. Azure regions that support Availability Zones have a minimum of three separate zones. Each availability zone consists of one or more data centers equipped with independent infrastructure power, network, and cooling. Availability zones are connected by a high-performance network with a round-trip latency of less than 2 milliseconds. For more information, see Overview of Availability Zones.

To protect your Virtual Machine Scale Sets from datacenter-level failures, you can create a scale set across Availability Zones. To use Availability Zones, your scale set must be created in a supported Azure region.

Design considerations for availability zones

Virtual Machine Scale Sets supports three zonal deployment models:

  • Zone redundant or zone spanning (recommended)
  • Zonal or zone aligned (single zone)
  • Regional

Zone redundant or zone spanning

A zone redundant or zone spanning scale set spreads instances across all selected zones, "zones": ["1","2","3"]. By default, the scale set performs a best effort approach to evenly spread instances across selected zones. However, you can specify that you want strict zone balance by setting "zoneBalance": "true" in your deployment. Each VM and its disks are zonal, so they are pinned to a specific zone. Instances between zones are connected by high-performance network with low latency. In the event of a zonal outage or connectivity issue, connectivity to instances within the affected zone may be compromised, while instances in other availability zones should be unaffected. You may add capacity to the scale set during a zonal outage, and the scale set adds more instances to the unaffected zones. When the zone is restored, you may need to scale down your scale set to the original capacity. A best practice would be to configure autoscale rules based on CPU or memory usage. The autoscale rules would allow the scale set to respond to a loss of the VM instances in that one zone by scaling out new instances in the remaining operational zones.

Spreading instances across availability zones meets the 99.99% SLA for instances spread across availability zones, and is recommended for most workloads in Azure.

Zonal or zone aligned (single zone)

A zonal or zone aligned scale set places instances in a single availability zone "zones": ['1']. Each VM and its disks are zonal, so they are pinned to a specific zone. This configuration is primarily used when you need lower latency between instances.

Regional

A regional Virtual Machine Scale Set is when the zone assignment isn't explicitly set ("zones"=[] or "zones"=null). In this configuration, the scale set creates Regional (not-zone pinned) instances and implicitly places instances throughout the region. There is no guarantee for balance or spread across zones, or that instances land in the same availability zone. Disk colocation is guaranteed for Ultra and Premium v2 disks, best effort for Premium V1 disks, and not guaranteed for Standard SKU (SSD or HDD) disks.

In the rare case of a full zonal outage, any or all instances within the scale set may be impacted.

Fault domains and availability zones

A fault domain is a fault isolation group within an availability zone or datacenter of hardware nodes that share the same power, networking, cooling, and platform maintenance schedule. VM instances that are on different fault domains are not likely to be impacted by the same planned or unplanned outage. You can specify how instances are spread across fault domains within a region or zone.

  • Max spreading (platformFaultDomainCount = 1)
  • Fixed spreading (platformFaultDomainCount = 5)
  • Fixed spreading aligned with storage disk fault domains (platformFaultDomainCount = 2 or 3, for regional deployments only)

With max spreading, the scale set spreads your VMs across as many fault domains as possible within each zone. This spreading could be across greater or fewer than five fault domains per zone. With static fixed spreading, the scale set spreads your VMs across the specified number of fault domains. If the scale set can't allocate to at least the specified fault domain count to satisfy the allocation request, the request fails.

We recommend deploying with max spreading for most workloads, as this approach provides the best spreading in most cases. If you need replicas to be spread across distinct hardware isolation units, we recommend spreading across Availability Zones and utilize max spreading within each zone.

Note

With max spreading, you only see one fault domain in the scale set VM instance view and in the instance metadata regardless of how many fault domains the VMs are spread across. The spreading within each zone is implicit.

Placement groups

Important

Placement groups only apply to Virtual Machine Scale Sets running in Uniform orchestration mode.

When you deploy a scale set, you can deploy with a single placement group per Availability Zone, or with multiple per zone. For regional (non-zonal) scale sets, the choice is to have a single placement group in the region or to have multiple in the region. If the scale set property called singlePlacementGroup is set to false, the scale set can be composed of multiple placement groups and has a range of 0-1,000 VMs. When set to the default value of true, the scale set is composed of a single placement group, and has a range of 0-100 VMs. For most workloads, we recommend multiple placement groups, which allows for greater scale. In API version 2017-12-01, scale sets default to multiple placement groups for single-zone and cross-zone scale sets, but they default to single placement group for regional (non-zonal) scale sets.

Note

If you use max spreading, you must use multiple placement groups.

Zone balancing

For scale sets deployed across multiple zones, you also have the option of choosing "best effort zone balance" or "strict zone balance." A scale set is considered "balanced" if each zone has the same number of VMs +\- 1 VM as all other zones for the scale set. For example:

  • A scale set with 2 VMs in zone 1, 3 VMs in zone 2, and 3 VMs in zone 3 is considered balanced. There is only one zone with a different VM count and it is only 1 less than the other zones.
  • A scale set with 1 VM in zone 1, 3 VMs in zone 2, and 3 VMs in zone 3 is considered unbalanced. Zone 1 has 2 fewer VMs than zones 2 and 3.

It's possible that VMs in the scale set are successfully created, but extensions on those VMs fail to deploy. These VMs with extension failures are still counted when determining if a scale set is balanced. For instance, a scale set with 3 VMs in zone 1, 3 VMs in zone 2, and 3 VMs in zone 3 is considered balanced even if all extensions failed in zone 1 and all extensions succeeded in zones 2 and 3.

With best-effort zone balance, the scale set attempts to scale in and out while maintaining balance. However, if for some reason zone balance isn't possible (for example, if one zone goes down, the scale set can't create a new VM in that zone), the scale set allows temporary imbalance to successfully scale in or out. On subsequent scale-out attempts, the scale set adds VMs to zones that need more VMs for the scale set to be balanced. Similarly, on subsequent scale in attempts, the scale set removes VMs from zones that need fewer VMs for the scale set to be balanced. With "strict zone balance", the scale set fails any attempts to scale in or out if doing so would cause unbalance.

To use best-effort zone balance, set zoneBalance to false. This setting is the default in API version 2017-12-01. To use strict zone balance, set zoneBalance to true.

Note

The zoneBalance property can only be set if the zones property of the scale set contains more than one zone. If there are no zones or only one zone specified, then zoneBalance property should not be set.

Create zone spanning or zonal scale sets

When you deploy a Virtual Machine Scale Set, you can choose to use a single Availability Zone in a region, or multiple zones.

You can create a scale set that uses Availability Zones with one of the following methods:

Use the Azure portal

The process to create a scale set that uses an Availability Zone is the same as detailed in the getting started article. When you select a supported Azure region, you can create a scale set in one or more available zones, as shown in the following example:

Create a scale set in a single Availability Zone

The scale set and supporting resources, such as the Azure load balancer and public IP address, are created in the single zone that you specify.

Use the Azure CLI

The process to create a scale set that uses an Availability Zone is the same as detailed in the getting started article. To use Availability Zones, you must create your scale set in a supported Azure region.

Add the --zones parameter to the az vmss create command and specify which zone to use (such as zone 1, 2, or 3).

az vmss create \
    --resource-group myResourceGroup \
    --name myScaleSet \
    --image <SKU Image> \
    --upgrade-policy-mode automatic \
    --admin-username azureuser \
    --generate-ssh-keys \
    --zones 1 2 3

It takes a few minutes to create and configure all the scale set resources and VMs in the zone(s) that you specify. For a complete example of a zone-redundant scale set and network resources, see this sample CLI script

Use Azure PowerShell

To use Availability Zones, you must create your scale set in a supported Azure region. Add the -Zone parameter to the New-AzVmssConfig command and specify which zone or zones to use (such as zone 1, 2, or 3).

New-AzVmss `
  -ResourceGroupName "myResourceGroup" `
  -Location "EastUS2" `
  -VMScaleSetName "myScaleSet" `
  -VirtualNetworkName "myVnet" `
  -SubnetName "mySubnet" `
  -PublicIpAddressName "myPublicIPAddress" `
  -LoadBalancerName "myLoadBalancer" `
  -UpgradePolicy "Automatic" `
  -Zone "1", "2", "3"

Use Azure Resource Manager templates

The process to create a scale set that uses an Availability Zone is the same as detailed in the getting started article for Linux or Windows.

{
  "type": "Microsoft.Compute/virtualMachineScaleSets",
  "name": "myScaleSet",
  "location": "East US 2",
  "apiVersion": "2017-12-01",
  "zones": [
        "1",
        "2",
        "3"
      ]
}

If you create a public IP address or a load balancer, specify the "sku": {"name":"Standard"} property to create zone-redundant network resources. You also need to create a Network Security Group and rules to permit any traffic. For more information, see Azure Load Balancer Standard Overview and Standard Load Balancer and Availability Zones.

Update scale set to add availability zones

You can modify a scale to expand the set of zones over which to spread VM instances. Expanding allows you to take advantage of higher zonal availability SLA (99.99%) versus regional availability SLA (99.95%). Or expand your scale set to take advantage of new availability zones that were not available when the scale set was created.

This feature can be used with API version 2023-03-01 or greater.

Expand scale set to use availability zones

You can update the scale set to scale out instances to one or more additional availability zones, up to the number of availability zones supported by the region. For regions that support zones, the minimum number of zones is 3.

Important

When you expand the scale set to additional zones, the original instances are not migrated or changed. When you scale out, new instances will be created and spread evenly across the selected availability zones. Data from the original instances are not migrated to the new zones. When you scale in the scale set, any regional instances will be priorized for removal first. After that, instances will be removed based on the scale in policy.

Expanding to a zonal scale set is done in 3 steps:

  1. Prepare for zonal expansion
  2. Update zones parameter on the scale set
  3. Add new zonal instances and remove original instances

Prepare for zonal expansion

Warning

This feature allows you to add zones to the scale set. You can't go back to a regional scale set or remove zones once they have been added.

In order to prepare for zonal expansion:

  • Check that you have enough quota for the VM size in the selected region to handle more instances.
  • Check that the VM size and disk types you are using are available in all the desired zones. You can use the Compute Resources SKUs API to determine which sizes are available in which zones
  • Validate that the scale set configuration is valid for zonal scale sets:
    • platformFaultDomainCount must be set to 1 or 5. Fixed spreading with 2 or 3 fault domains isn't supported for zonal deployments.
    • Capacity reservations are not supported during zone expansion. Once the scale set is fully zonal (no more regional instances), you can add a capacity reservation group to the scale set.
    • Azure Dedicated Host deployments are not supported.

Update the zones parameter on the scale set

Update the scale set to change the zones parameter.

  1. Navigate to the scale set you want to update
  2. On the Availability tab of the scale set landing page, find the Availability zone property and press Edit
  3. On the Edit Location dialog box, select the desired zone(s)
  4. Select Apply

Add new zonal instances and remove original instances

Manually scale out and in

Update the capacity of the scale set to add more instances. The new capacity should be set to the original capacity plus the number of new instances. For example, if your scale set had 5 regional instances and you would like to scale out so that you have 3 instances in each of 3 zones, you should set the capacity to 14.

You can update the zones parameter and the scale set capacity in the same ARM template or REST API call.

When you are satisfied that the new instances are ready, scale in your scale set to remove the original regional instances. You can either manually delete the specific regional instances, or scale in by reducing the scale set capacity. When scaling in via reducing scale set capacity, the platform will always prefer removing the regional instances, then follow the scale in policy.

Known issues and limitations

  • The original instances are not migrated to the newly added zones. Your workload must handle any required data migration or replication.

  • Scale sets running Service Fabric RP or Azure Kubernetes Service are not supported.

  • You can't remove or replace zones, only add zones

  • You can't update from a zone spanning or zonal scale set to a regional scaleset.

  • platformFaultDomainCount must be set to 1 or 5. Fixed spreading with 2 or 3 fault domains isn't supported for zonal deployments.

  • Capacity reservations are not supported during zone expansion. Once the scale set is fully zonal (no more regional instances), you can add a capacity reservation group to the scale set.

  • Azure Dedicated Host deployments are not supported

Next steps

Now that you have created a scale set in an Availability Zone, you can learn how to Deploy applications on Virtual Machine Scale Sets or Use autoscale with Virtual Machine Scale Sets.