Proximity placement groups
Applies to: ✔️ Linux & Windows VMs ✔️ Flexible & Uniform scale sets
To get VMs as close as possible, achieving the lowest possible latency, you should deploy them within a
proximity placement group.
A proximity placement group is a logical grouping used to make sure that Azure compute resources are
physically located close to each other.
Proximity placement groups are useful for workloads where low latency is a requirement.
You can also move an existing resource into a proximity placement group. When moving a resource into a
proximity placement group, you should stop (deallocate) the asset first.
P.P.G Cannot applied on cross region OR Availability Zone.
You can use different types of VM Family.
To achieve more low latency you can also use proximity placement group with accelerated networking.
Azure Dedicated Hosts (Dedicated Hosts Group)
Dedicated hosts are the same physical servers used in azur’s data centers, provided as a resource.
You can provision dedicated hosts within a region, availability zone, and fault domain. Then, you can place
VMs directly into your provisioned hosts, in whatever configuration best meets your needs.
A host group is a resource that represents a collection of dedicated hosts.
A host is a resource, mapped to a physical server in an Azure data center. A host has a SKU describing
which VM sizes can be created. You can mix multiple VMs of different sizes within a single host as long as
they are of the same size series.
Support VM and VMSS.
Reserving the entire host provides:
• Hardware isolation at the physical server level. No other VMs will be placed on your hosts.
Dedicated hosts are deployed in the same data centers and share the same network and underlying
storage infrastructure.
• Control over maintenance events initiated by the Azure platform. With dedicated hosts, you can opt-
in to a maintenance window to reduce the impact to your service.
• you can bring your own licenses for Windows and SQL to Azure.
For high availability, you should deploy multiple VMs, spread across multiple hosts (minimum of 2).
When creating a VM in Azure, You can also use the option to automatically place your VMs on existing
hosts, within a host group.
limitations when using automatic VM placement:
• You will not be able to redeploy your VM.
• You will not be able to use Lsv2, NVasv4, NVsv3, Msv2, or M-series VMs
Virtual machine scale set in a dedicated host group:
• Automatic VM placement needs to be enabled.
• The host group and the scale set must be using the same availability zone.
• The fault domain count for the host group level should match the fault domain count for
your scale set. The Azure portal lets you specify max spreading for your scale set, which sets
the fault domain count of 1.
• Overprovisioning is disabled by default. You can enable overprovisioning.
• Do not use proximity placement groups for co-location
Maintenance control:
provides customers with an option to skip regular platform updates scheduled on their dedicated hosts,
then apply it at the time of their choice within a 35-day rolling window. Within the maintenance window,
you can apply maintenance directly at the host level, in any order. Once the maintenance window is over,
Microsoft will move forward and apply the pending maintenance to the hosts.
The default quota limit is 3000 vCPUs, per region.
VM size family quota. It can vary for Different VM Series.
Once a Dedicated host is provisoned, you can't change the size or type.
