Confidential Computing extends encryption to data in use, protecting sensitive workloads from hypervisor-level and certain physical access scenarios.
What is Confidential Computing?
Traditional cloud security encrypts data at rest (storage) and in transit (network), but data must typically be decrypted for processing. Confidential Computing closes that gap: it encrypts data in use, while it sits in memory during processing.
Depending on the machine series, Confidential VMs use different hardware technologies for memory encryption: AMD SEV or the enhanced SEV-SNP on N2D instances, Intel TDX on C3 instances, and AMD SEV on C4D instances (currently in preview). Encryption keys are managed directly within the CPU.
Core Features
- Hardware-based encryption: AMD SEV/SEV-SNP or Intel TDX encrypt VM memory at the CPU level
- Multiple supported machine series: N2D, C3, and C4D (preview)
- Remote attestation: Cryptographically verify the security configuration of a workload
- Low overhead: Google states the overhead for AMD SEV ranges from negligible to minimal
- Confidential GKE Nodes: Run Kubernetes workloads on confidential nodes
- Confidential Space: A trusted environment for multi-party computation without exposing raw data
Typical Use Cases
Financial Services Data Processing
Banks and financial institutions process sensitive customer data, trading algorithms, and risk models. Confidential Computing adds protection for this data even during computation.
Healthcare Analytics
Healthcare organizations analyze patient data for research and clinical decision support. Confidential Computing allows processing of protected health information with an additional layer of protection against unauthorized access.
Multi-Party Data Collaboration
Multiple organizations can jointly analyze combined datasets without exposing their raw data to each other. Confidential Space provides a trusted, attested execution environment for this kind of computation.
Confidential VMs vs. Shielded VMs
| Feature | Shielded VMs | Confidential VMs |
|---|---|---|
| Secure Boot | Yes | Yes |
| vTPM | Yes | Yes |
| Integrity Monitoring | Yes | Yes |
| Memory encryption | No | Yes |
| Supported machine series | Broad range | N2D, C3, C4D (preview) |
Benefits
- Additional security layer: Protection against hypervisor-level access and certain physical attack scenarios
- No code changes: Confidential VMs work with existing applications
- Hardware-based: AMD SEV/SEV-SNP or Intel TDX handle encryption in the CPU
- Attestation: Cryptographic proof of the security configuration is available
Integration with innFactory
As a certified Google Cloud Partner, innFactory helps you implement Confidential Computing for sensitive workloads. We assess which applications benefit most from confidential environments, design architectures that leverage Confidential VMs and GKE, and help you demonstrate compliance with regulatory requirements.
Available Tiers & Options
Confidential VMs (AMD SEV / SEV-SNP, Intel TDX)
- No code changes required
- Full VM memory encryption
- Minimal to negligible performance overhead for AMD SEV
- Limited to specific machine series
Confidential GKE Nodes
- Kubernetes-native integration
- Node-level encryption
- Works with existing workloads
- Limited to supported node types
Confidential Space
- Multi-party computation
- Attestation-based trust
- Data clean rooms
- Requires workload redesign
Typical Use Cases
Technical Specifications
Frequently Asked Questions
What is Confidential Computing?
Confidential Computing encrypts data while it is being processed in memory, not just at rest or in transit. Encryption happens at the hardware level, for example via AMD SEV/SEV-SNP or Intel TDX, so that data is protected even from access at the hypervisor level or certain physical attack scenarios.
How much performance overhead does Confidential Computing add?
According to Google, the performance difference between an AMD SEV Confidential VM and a standard Compute Engine VM ranges from negligible to minimal. The exact impact depends on the workload and the chosen technology.
Do I need to modify my applications?
No. Confidential VMs are transparent to applications. You select a supported machine type and configuration, and memory encryption happens automatically at the hardware level. Existing applications run without code changes.
Which machine types support Confidential Computing?
Confidential VMs are available across several machine series, including N2D with AMD SEV or SEV-SNP, C3 with Intel TDX, and C4D with AMD SEV currently in preview. The full, current list of supported configurations is maintained in Google's documentation.
How does Confidential Computing differ from encryption at rest?
Encryption at rest protects stored data, and encryption in transit protects network data. Confidential Computing protects data during processing, in the memory where your application runs. Together, they cover the full data lifecycle.
Note: All product information on this page has been compiled with care, but is provided without guarantee and may be outdated or incomplete. Cloud services evolve rapidly — features, pricing, SLAs, and availability change frequently. Authoritative and up-to-date information can only be found on the official product page of Google Cloud (official documentation). This page does not represent an offer by Google Cloud.
