A guide to dual-region storage in Google Cloud Storage, now available in Frankfurt, London, Zurich and Belgium

In today's digital landscape, data reliability, availability, and performance are paramount. Businesses and organizations rely on cloud storage solutions that not only ensure that data is safe, but that also provide uninterrupted access and low latency. Cloud Storage is a robust solution to these challenges, offering three distinct types of bucket location types, each with their own pros and cons: single region, dual-region, and multi-region.

This is especially timely, because we continue to add Cloud Storage dual-region pairings to help businesses navigate and streamline European data compliance rules. Today, we’re excited to announce three new dual-region pairings: Frankfurt/Zurich, Frankfurt/London, and Belgium/London. By enabling data replication across geographically distinct, yet proximate locations, Cloud Storage offers data resilience and sovereignty while helping organizations adhere to stringent regulatory requirements.

Read on as we explore what dual-region storage is, how to use it effectively, and its advantages.

Understanding dual-region storage

So, what exactly is dual-region storage, and how does it work? In Cloud Storage, you need to set a geographic location for storing the object data when you create a bucket. Dual-region buckets store data redundantly across two geographically distant regions within a single continent while presenting as a single bucket (namespace) with strong consistency (more on that below). This setup enhances data resilience by ensuring that even if one entire region becomes unavailable due to a disaster or maintenance, the data remains accessible from the secondary region with no downtime. This can be especially attractive for customers who want to split their workload across the two regions as part of a business continuity / disaster recovery architecture.

Active-active redundancy across two regions

Unlike traditional storage models that take an active-passive approach, implementing  "primary" and "secondary" geographic locations, Cloud Storage dual-regions use an active-active architecture based on a single geo-redundant bucket. This simplifies the disaster recovery process by eliminating the need to replicate data from one bucket to another, manually failover to a secondary bucket in the case of primary region downtime, or manually failback to the primary bucket when primary region availability is restored. In the event of a regional outage, Cloud Storage dual-region buckets continue serving all data that has been replicated across regions as required, for a recovery time objective (RTO) of zero. Temporary regional failures are normally invisible to users.

As a best practice, consider load balancing the rest of your workload across the two regions making up your dual-region bucket to take advantage of the active-active storage architecture. If you take this approach, your workload will continue to operate uninterrupted in the case of either region having an outage without the need for manual intervention. Alternatively, dual-region storage will still allow you to run the rest of your use case in a traditional primary-secondary mode across the two regions if an active-active architecture isn’t possible for whatever reason. In this case, if your primary region has an outage, you will need to fail over your compute and other resources to the secondary region where your dual-region bucket will already be available to serve your data.

Strong consistency

Dual-region buckets benefit from the same strong consistency guarantees that are available for single-region and multi-region bucket . Strong consistency is key to making the active-active topology possible—no matter when or where data is written to a dual-region bucket, you will see the most recent view of your data, without having to worry about more than one regional bucket getting out of sync with each other. Strong consistency is particularly important for ensuring that all relevant data is available for analysis to avoid inaccurate or incomplete results, for example in data analytics scenarios.

Cloud Storage dual-region buckets achieve their geo-redundancy through asynchronous data replication. During this process, newly written objects are initially protected as regional objects with redundancy across availability zones within a single region. Within a specified Recovery Point Objective (RPO) time window, the service replicates these objects to a second region, transforming them into geo-redundant objects. The duration of object replication between the two regions is determined by the bucket's replication settings

• Default replication provides redundancy across regions for 99.9% of newly written objects within a RPO target of one hour and 100% of newly written objects with an RPO of 12 hours. Newly written objects include uploads, rewrites, copies, and compositions. As you might expect from its name, this type of replication is the default for multi-regions and dual-regions.

• Turbo replication is a premium feature that provides an industry-leading 15-min RPO. It is backed by the Cloud Storage Service Level Agreement (SLA) to replicate 100% of newly written objects to both regions in the dual-region within 15 minutes, regardless of object size. By providing faster redundancy across regions for data in your dual-region buckets, turbo replication reduces the risk of data loss and helps support uninterrupted service following a regional outage.

Benefits of dual-region storage

1. Improved availability and reliability: Dual-region storage provides significantly higher availability than regional storage. If one region experiences downtime or is undergoing maintenance, data is transparently served from the other region.

2. Supports active-active and active-passive disaster recovery architectures: Overall workflows are best load balanced across the two regions of a dual-region in an active-active topology, made possible by the strong consistency of dual-region storage. However, dual-regions can equally well support primary-secondary topologies if that is your preferred approach.

3. Ease of management: Dual-region data replication and any necessary regional failover and failback is performed automatically without the need for manual intervention. This supports an RTO of zero,  improving business continuity and disaster recovery.

4. Higher performance: Workloads experience better performance when they access data from the region closest to them. Moreover, dual-region buckets can support terabits per second of read bandwidth, comparable to single-region buckets.

5. Compliance and data sovereignty: Dual-region storage helps organizations comply with data sovereignty regulations by keeping data within specific geographic boundaries while still ensuring redundancy and availability. For example, you can use dual-region storage to address industry-specific requirements regarding the physical distance between copies of the data.

In short, dual-region storage in Google Cloud is a powerful way to enhance data resilience, availability, and performance. By leveraging redundant storage across geographically separated regions, businesses can mitigate risks, improve reliability, and comply with regulatory requirements. Whether you're running mission-critical applications or managing sensitive data, dual-region storage offers a robust foundation for your cloud infrastructure needs on Google Cloud. Embrace the advantages of dual-region storage today and take your data reliability to the next level.

Refer to Cloud Storage dual-region documentation to get started.