Confidential VM instances are a type of [Compute Engine](https://docs.cloud.google.com/compute/docs)
virtual machine. They use hardware-based memory encryption to help ensure that
your data and applications can't be read or modified while in use.

Confidential VM instances offer the following benefits:

- **Isolation**: Encryption keys are generated by---and reside solely
  in---dedicated hardware, inaccessible to the hypervisor.

- [**Attestation**](https://docs.cloud.google.com/confidential-computing/confidential-vm/docs/attestation-overview): You can verify the
  identity and the state of the VM, to make sure that key components haven't
  been tampered with.

This type of hardware isolation and attestation is known as a
*Trusted Execution Environment* (TEE).

You can
[enable the Confidential VM service](https://docs.cloud.google.com/confidential-computing/confidential-vm/docs/create-a-confidential-vm-instance)
whenever you create a new VM instance.

## Confidential Computing technologies

When setting up a Confidential VM instance, the type of Confidential Computing
technology that's used is based on the
[machine type and CPU platform you choose](https://docs.cloud.google.com/confidential-computing/confidential-vm/docs/supported-configurations).
When choosing a Confidential Computing technology, make sure it fits your
performance and [cost](https://docs.cloud.google.com/confidential-computing/confidential-vm/pricing) needs.

### AMD SEV

AMD Secure Encrypted Virtualization (SEV) on Confidential VM offers hardware-based
memory encryption through the AMD Secure Processor, and boot-time attestation
through Google's vTPM.

AMD SEV offers high performance for demanding computational tasks. The
performance difference between an SEV Confidential VM and a standard
Compute Engine VM can range from nothing to minimal, depending on the
workload.

Unlike other Confidential Computing technologies on Confidential VM, AMD SEV
machines that use the N2D and C3D machine types support live migration.

Read the
[AMD SEV whitepaper](https://www.amd.com/content/dam/amd/en/documents/epyc-business-docs/white-papers/memory-encryption-white-paper.pdf).

### AMD SEV-SNP

AMD Secure Encrypted Virtualization-Secure Nested Paging (SEV-SNP) expands on
SEV, adding hardware-based security to help prevent malicious hypervisor-based
attacks like data replay and memory remapping. Attestation reports can be
requested at any time directly from the AMD Secure Processor.

Read the
[AMD SEV-SNP whitepaper](https://www.amd.com/content/dam/amd/en/documents/epyc-business-docs/white-papers/SEV-SNP-strengthening-vm-isolation-with-integrity-protection-and-more.pdf).

### Intel TDX

Intel Trust Domain Extensions (TDX) creates an isolated trust domain (TD) within
a VM, and uses hardware extensions for managing and encrypting memory.

Intel TDX augments defense of the TD against limited forms of attacks that use
physical access to the platform memory, such as offline, dynamic random access
memory (DRAM) analysis and active attacks of DRAM interfaces. These attacks
include capturing, modifying, relocating, splicing, and aliasing memory
contents.

Read the
[Intel TDX whitepaper](https://www.intel.com/content/www/us/en/developer/tools/trust-domain-extensions/documentation.html).

### NVIDIA Confidential Computing

NVIDIA Confidential Computing extends the security benefits of a hardware-based trusted
execution environment (TEE) to attached NVIDIA GPUs. Sensitive data in
GPU-accelerated AI and ML workloads is encrypted in-use, providing isolation
from the hypervisor.

To provide a comprehensive confidential environment for both CPU and GPU
workloads, NVIDIA Confidential Computing is integrated with CPU-based confidential computing
technologies such as AMD SEV or Intel TDX.

For details on which machine series, GPU models, and CPU platforms support
NVIDIA Confidential Computing, see
[Supported configurations](https://docs.cloud.google.com/confidential-computing/confidential-vm/docs/supported-configurations).

Read the
[NVIDIA H100 Tensor Core GPU Architecture](https://resources.nvidia.com/en-us-hopper-architecture/nvidia-h100-tensor-c)
and the
[RTX PRO 6000 Blackwell GPU Architecture](https://docs.nvidia.com/nvidia-secure-ai-with-blackwell-and-hopper-gpus-whitepaper.pdf)
whitepapers.

> [!NOTE]
> **Note:** NVIDIA HGX B200 and NVIDIA GB300 NVL72 access is available to specific eligible customers. To check your eligibility and to request access to these systems, contact your dedicated account team.

## Confidential VM services

In addition to Compute Engine, the following Google Cloud services make
use of Confidential VM:

- [Confidential Google Kubernetes Engine Nodes](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/confidential-gke-nodes)
  enforce the use of Confidential VM for all your GKE nodes.

- [Confidential Space](https://docs.cloud.google.com/confidential-computing/confidential-space/docs/confidential-space-overview) uses
  Confidential VM to let parties share sensitive data with a mutually agreed upon
  workload, while they retain confidentiality and ownership of that data.

- [Managed Service for Apache Spark Confidential Compute](https://docs.cloud.google.com/dataproc/docs/concepts/configuring-clusters/confidential-compute)
  features Managed Service for Apache Spark clusters that use Confidential VM.

- [Dataflow Confidential VM](https://docs.cloud.google.com/dataflow/docs/reference/service-options)
  features Dataflow worker Confidential VM instances.

## What's next

Read about Confidential VM
[supported configurations](https://docs.cloud.google.com/confidential-computing/confidential-vm/docs/supported-configurations).