A robust backup strategy is essential for maintaining cyber resilience.
Google Cloud offers Backup and DR, a managed solution designed for
protecting and restoring workloads in Google Cloud environments. This service
provides a unified platform for the centralized management, monitoring, and
reporting of daily backup activities, and it integrates smoothly with existing
automation tools like APIs and Terraform.

This page outlines architectural recommendations and configuration best practices
for Google Cloud Backup and DR. These guidelines are intended to help
you construct a sturdy defense against cyber threats, with a particular focus on
mitigating the risks of unauthorized data exfiltration and ransomware attacks.

## Plan the backup architecture

A well-architected backup environment is the first step to limiting the damage
that can be caused in the event of a security breach.

### Implement a centralized model

Using a centralized model (similar to a hub-and-spoke model) lets you
isolate your backups from your production systems, thereby limiting the impact
of possible cyber attacks while also making it easier to manage and govern your
data protection.

- The Hub (Backup Project): Create a dedicated, restricted Google Cloud project
  for your Backup Vaults to store your backed up data. This project also becomes
  your central management hub for managing your backups.

- The Spokes (Workload Projects): These include the projects that contain your
  production and development workloads, which need to be protected. This isolation
  ensures that even if a 'Spoke' project is compromised, the attacker does not
  have inherent access to the 'Hub' project where the recovery data is stored.

For more details about centralized and decentralized backup strategies, see
[Overview of backup vaults](https://docs.cloud.google.com/backup-disaster-recovery/docs/concepts/backup-vault).

## Manage identity and access

Identity is often referred to as 'the new perimeter', referring to security
perimeters going beyond just network firewalls. Protecting backup administrative
credentials and restricting access is one of the most critical steps in
preventing ransomware from deleting your backups.

### Apply minimal access to backup vaults

Apply the principle of least privilege strictly to your backup projects:

- Separate roles: Ensure that the user who can trigger a backup is not the same
  user who has the authority to delete a backup or modify retention policies.

- Restrict Backup Admin roles: Limit the number of users with Cloud Backup and DR
  Admin permissions. Most users should only have Viewer or Operator roles which
  should suffice for them to perform their duties.

[Control access to Backup and DR with IAM](https://docs.cloud.google.com/backup-disaster-recovery/docs/access-control#roles)
throws more light on the Identity and Access Management (IAM) roles and permissions that
are required by Backup and DR.

### Manage service agent permissions

Backup and DR uses a service agent to perform cross-project operations.

Ensure the service agent has the minimum required permissions in the 'Spoke'
projects to capture and recover data.

Periodically audit the service agent's permissions to ensure no 'privilege
creep' has occurred.

## Configure core backup capabilities

Google Cloud Backup provides various built-in features that can help you ensure
that your backup data cannot be deleted, altered or bypassed.

### Enable immutability and lock settings

Immutable Backups: Configure your backup vaults to specify and lock a minimum
retention period. This prevents any users, including those with administrative
privileges, from deleting backup data before the intention retention period
expires.

Org Policies: Use Google Cloud Organization Policies to enforce the use of
immutable vaults across all projects. To learn more about using the
Organization Policy Service service to implement custom constraints for specific operations,
see [Creating custom constraints for Backup and DR](https://docs.cloud.google.com/backup-disaster-recovery/docs/customconstraints).

### Enable customer-managed encryption keys

Using customer-managed encryption keys (CMEK) for backups lets you protect your
backup data using a cryptographic key that is managed by you through the
Cloud Key Management Service (Cloud KMS).

When you use CMEK, you manage the key in Cloud KMS, and you
can control who can access it by managing IAM permissions on the
key. You can configure CMEK on a backup vault only when you
create it. You cannot enable or disable CMEK on an existing backup vault.

Learn more about CMEK for encrypting your backups in
[Customer-managed encryption keys (CMEK)](https://docs.cloud.google.com/backup-disaster-recovery/docs/concepts/cmek).

### Use multi-region vaults

Google Cloud Backup multi-region vaults store data across several geographic
areas. This provides enhanced security and ensures backup availability during
unforeseen events.

Benefits of multi-region backup vaults include:

- Continuous access: Maintain data recoverability during regional disruptions like power outages or natural disasters.
- Business continuity: Support on-demand recoveries to ensure reliable business operations.
- Data security: Retain all core security advantages provided by standard backup vaults.

To learn more about support for multi-region backup vaults, see [Overview of backup vaults](https://docs.cloud.google.com/backup-disaster-recovery/docs/concepts/backup-vault#multi-regions).

### Set access restrictions

Restricting vault access to and from its organization or project can help
minimize chances of unauthorized access to your backup data. Note that this
setting is configured during vault creation and cannot be changed later.

Refer to
[Backup vault access restriction](https://docs.cloud.google.com/backup-disaster-recovery/docs/concepts/backup-vault#access-restrictions)
for more details.

## Configure integrated security alerts

Backup and DR integrates with Security Command Center and Google Security Operations to
provide visibility into high-risk activities using alerts surfaced within those
platforms.

By leveraging Google SecOps and Security Command Center alongside
Backup and DR, you are able to:

- Get immediate notifications regarding critical actions, such as when workload protection is disabled.
- Conduct threat investigations and pinpoint specific backup resources that may be impacted.
- Consolidate backup-related threats into cases to facilitate efficient and structured mitigation.

[Learn more about Security Command Center and Google SecOps for Backup and DR](https://docs.cloud.google.com/backup-disaster-recovery/docs/monitor-reports/alerts-scc).

## Use Google Cloud platform guardrails

Use Google Cloud security features to add additional layers of protection
to your backup infrastructure.

### Use project liens

To protect your backup vault from accidental deletion, we strongly recommend
applying a project lien to the Google Cloud project where the vault will
reside. A project lien prevents the project---and your backups---from being
deleted unexpectedly. For more information, see [Protect projects with liens](https://docs.cloud.google.com/resource-manager/docs/project-liens).

To secure your project liens against unauthorized removal, use
Privileged Access Manager (PAM) to enforce multi-party approval. For more information,
see [Protect project liens by using Privileged Access Manager](https://docs.cloud.google.com/backup-disaster-recovery/docs/configuration/project-liens-multi-party-approval).

> [!IMPORTANT]
> **Important:** Beginning November 1, 2026, Backup and DR will automatically apply a project-level lien to any project containing a backup vault with backups protected by enforced retention.

### Use network constraints and VPC Service Controls

VPC Service Controls (VPC-SC): Define a service perimeter around your backup
projects to prevent data exfiltration.

Restrict External IPs: Ensure that backup management appliances don't have
public IP addresses.

## Summary of best practices

The following table summarizes the best practices recommended in this document:

| Topic | Task |
|---|---|
| Plan the backup architecture | Implement a centralized model to isolate backups from production systems. |
| Manage identity and access | Apply the principle of least privilege to backup vaults. Manage and audit service agent permissions. |
| Configure core backup capabilities | Enable immutability and lock settings to prevent premature data deletion. Enable customer-managed encryption keys (CMEK) for data protection. Use multi-region vaults for enhanced availability and security. Set access restrictions on backup vaults. |
| Configure integrated security alerts | Integrate with Security Command Center and Google SecOps for threat visibility. |
| Use Google Cloud platform guardrails | Use project liens to prevent accidental or malicious project deletion. Use network constraints and VPC Service Controls to prevent data exfiltration. |

## What's next

- Learn more about [centralized and decentralized backup strategies](https://docs.cloud.google.com/backup-disaster-recovery/docs/concepts/backup-vault).
- Review [IAM roles and permissions for Backup and DR](https://docs.cloud.google.com/backup-disaster-recovery/docs/access-control#roles).
- See how to use the [Organization Policy service for custom constraints](https://docs.cloud.google.com/backup-disaster-recovery/docs/customconstraints).
- Read about [using CMEK for encrypting backups](https://docs.cloud.google.com/backup-disaster-recovery/docs/concepts/cmek).
- Learn about [monitoring and alerts with Security Command Center](https://docs.cloud.google.com/backup-disaster-recovery/docs/monitor-reports/alerts-scc).