The *tenancy* of a virtual machine (VM) instance indicates whether the VM shares
its Compute Engine server with VMs from other Google Cloud projects. If a VM
shares its Compute Engine server with VMs from other
Google Cloud projects, it is a *multi-tenant* VM. If a VM doesn't share its
Compute Engine server with VMs from other projects, it is a
*sole-tenant* VM.

VMs are multi-tenant by default. After you create a VM, Compute Engine
places it on a multi-tenant server. Compute Engine allows users in
other
[Google Cloud projects](https://docs.cloud.google.com/resource-manager/docs/creating-managing-projects)
to create VMs on that server.

[Sole-tenancy](https://docs.cloud.google.com/compute/docs/nodes/sole-tenant-nodes) is suited for workloads
that require exclusive access to a Compute Engine server. Sole-tenancy
lets you have exclusive access to a *sole-tenant node*, which is a
Compute Engine server that is dedicated to hosting VMs from only your
Google Cloud projects.

## Pricing considerations

For multi-tenant VMs, you pay for the type of machine that you create the VM on,
so [pricing for multi-tenant VMs](https://cloud.google.com/compute/vm-instance-pricing) is based on that
VM's [machine family](https://docs.cloud.google.com/compute/docs/machine-resource).

For sole-tenant VMs, you pay for all of the hardware resources on the server
because you are reserving an entire physical server. [Pricing for a sole-tenant
node](https://cloud.google.com/compute/sole-tenant-pricing) is based on the price of the [sole-tenant
node type](https://docs.cloud.google.com/compute/docs/nodes/sole-tenant-nodes#node_types) that you specify
when you create a [sole-tenant node
template](https://docs.cloud.google.com/compute/docs/nodes/sole-tenant-nodes#node_templates). This price
includes a premium because you are reserving the entire physical server.

## Workload considerations

For typical workloads, [create multi-tenant
VMs](https://docs.cloud.google.com/compute/docs/instances/create-start-instance). However, if you have
workloads with any of the following characteristics, consider [using sole-tenant
nodes](https://docs.cloud.google.com/compute/docs/nodes/provisioning-sole-tenant-vms):

- Gaming workloads with specific computing performance requirements.
- Finance or healthcare workloads with security and compliance requirements.
- Windows workloads with licensing requirements.
- Machine learning, data processing, or image rendering workloads. For these workloads, consider [reserving GPUs](https://docs.cloud.google.com/compute/docs/nodes/provisioning-sole-tenant-vms).
- Workloads requiring increased input/output operations per second (IOPS) and decreased latency, or workloads that use temporary storage in the form of caches, processing space, or low-value data. For these workloads, consider [reserving Local SSDs](https://docs.cloud.google.com/compute/docs/nodes/provisioning-sole-tenant-vms).

For more information about workloads that might benefit from using sole-tenancy,
see [Workload considerations for sole-tenant
nodes](https://docs.cloud.google.com/compute/docs/nodes/sole-tenant-nodes#workload_considerations).

## Maintenance event considerations

For maintenance events on multi-tenant servers, Compute Engine [live
migrates](https://docs.cloud.google.com/compute/docs/instances/live-migration-process) VMs to another server in the
same zone.

For maintenance events on sole-tenant nodes, Compute Engine migrates
VMs according to how you configure the [maintenance
policy](https://docs.cloud.google.com/compute/docs/nodes/sole-tenant-nodes#maintenance_policies) on the
[sole-tenant node
group](https://docs.cloud.google.com/compute/docs/nodes/sole-tenant-nodes#node_groups_and_vm_provisioning).