Data residency overview

Overview

This page explains how to use Cloud SQL to enforce data residency requirements.

Data residency refers to the physical location of data and the local regulations that govern how you store, encrypt, and access that data. As countries' data protections and privacy regulations evolve, it's increasingly important that you understand how to follow local data residency requirements and protect your users' data.

In a traditional on-premises environment, various components integrate and handle data residency. For example, a company can host a tokenization gateway as a Cloud Access Security Broker (CASB) to secure application data before it's transmitted overseas.

Google Cloud and its services, including Cloud SQL, integrate to help you handle data residency by controlling the location of your data and access to that data—by Google or by anyone.

Data residency in cloud computing

The following lists some data residency issues about which you should know:

  • If a company's cloud administrators don't know the physical location of the data, then they don't know the local regulations. To be able to research the data residency policies for each location, administrators need to know where the data centers are.
  • The company's cloud administrators and providers can use service-level agreements (SLAs) to establish allowed locations. However, what if you have to store your data in a region that differs from the terms of the SLA?
  • Users must ensure that their data, and all of the services and resources that they use in their cloud projects, follow the data residency regulations of the host country.
    • What do you do if you want to decide where to store your data and encryption keys?
    • What happens if you want to determine the locations where users can access your data?

Google Cloud services, including Cloud SQL, address some of these issues by letting you do the following:

  • Set the storage location of your data. You can select the region when you create your Cloud SQL instance, or you can edit the region by editing an existing instance.
  • Use the cross-region read replica feature for Cloud SQL to help you meet data residency standards for a designated region.
  • Control the network locations where users can access data, as well as control cloud administrators' access to this data.

There are three areas where Cloud SQL can help you meet the challenges of data residency:

Store data

Data residency involves storing personally identifiable information (PII) within a particular region, where that data is processed in accordance with the regulations of that region.

To store data, you must meet a country's legal and regulatory demands, such as data locality laws. For example, a country might mandate that any government-related data be stored in that country. Or, a company might be contractually obligated to store data for some of their customers in a different country. Therefore, a company has to meet the data residency requirements of the country where the data is stored.

With Google Cloud, you can configure where your data is stored, including on backups. This includes letting you choose the regions where you store your data. When you choose to configure resources in these regions for Cloud SQL, Google stores your data at rest only in these regions, according to our Service Specific Terms. You can select the region when you create your instance, or you can edit the region by editing an existing instance.

For more information about backup locations, see Custom backup locations.

You can use organizational policy constraints to enforce data residency requirements at the organization, project, or folder level. These constraints let you define the Google Cloud locations where users can create resources for supported services. For data residency, you can limit the physical location of a new resource with the resource locations constraint. You can also fine-tune policies for a constraint to specify multi-regions such as asia and europe, or regions such as us-east1 or europe-west1, as allowed or denied locations.

If there's an outage in a Google Cloud region, then you can avoid impact by using the cross-region read replica feature for Cloud SQL. As part of creating a read replica, you must choose the region for the replica. Make sure that you choose a region that complies with data residency regulations. As a result, the data residency standards are met in the selected region.

When you create the replica, you create a copy of your primary database instance that reflects changes to the primary in near real time. If there's a failure, then you can promote the read replica to a primary instance.

VPC Service Controls help you enforce data residency by letting you restrict the use of Cloud SQL APIs to import and export data using either the Cloud SQL Admin API or the Cloud Storage API. This restriction helps ensure that data stays within the network locations you've selected. Using VPC Service Controls, you create a service perimeter that defines the virtual boundaries from which a service can be accessed, preventing data from being moved outside those boundaries. You can enforce this constraint even if the user is authorized according to your Google Cloud IAM policy.

Encrypt data

Google Cloud services, including Cloud SQL, encrypt customer content at rest and in transit using various encryption methods. Encryption is automatic, and no customer action is required.

Cloud SQL also lets you add another layer of encryption to data using customer-managed encryption keys (CMEK). CMEK are intended for organizations that have sensitive or regulated data that requires them to manage their own encryption keys. The CMEK feature lets you use your own cryptographic keys for data at rest in Cloud SQL. After you add CMEK, whenever an API call is made, Cloud SQL uses your keys to access data.

If you want to store your CMEK in the regions where you deploy your services, then you can use Cloud Key Management Service (Cloud KMS). You set the location of the key when you create the key. Or, to store those keys inside a physical hardware security module (HSM) that's located in the region that you choose, use Cloud HSM.

Another way to choose where you want to store your CMEK is to use a third-party product. To store and manage keys in a third-party key management product that's deployed outside of Google's infrastructure, you can use Cloud External Key Manager (Cloud EKM).

Access data

With Cloud SQL, you can control which users can access your data.

To control Google's access of support and engineering personnel, use Access Approval. Access Approval lets you require Google employees to get your explicit approval before they access your data or configurations on Google Cloud (for exclusions, see Access Approval exclusions).

Access Approval complements the visibility provided by Access Transparency, which generates near-real-time audit logs when Google administrators interact with your data. The audit logs include the office location of the administrator and the reason for the access. You can also enforce specific attributes for administrators who have access to your data or configurations—including their geographic region and other compliance-relevant attributes.

Key Access Justifications integrate with Cloud KMS and Cloud EKM. Each time one of your keys is requested to encrypt or decrypt data, Key Access Justifications provides a detailed justification, along with a mechanism for you to approve or deny key access using an automated policy that you set.

Using Access Approval, Access Transparency, and Key Access Justifications with Cloud KMS and Cloud EKM, you can deny Google the ability to decrypt your data. As a result, you're the arbiter of access to your data.

What's next