Choosing Primary and Secondary Regions in Snowflake for Business Continuity

Natural disaster. Errant code. Regional downtime. In the event of such disruptions, it can be crucial to have a secondary region for recovery to ensure business continuity and data durability. 

Snowflake Database Replication enables customers to replicate databases and keep them synchronized across multiple accounts in different regions and for different cloud providers. Customers can synchronize changes to a secondary region or cloud provider, ensuring data durability and availability at all times. Snowflake Database Failover comes into play in the event of a massive outage that disrupts the cloud services in a given region or for a specific cloud provider; it allows customers to fail over databases to an available region or cloud provider to continue business operations.

Figure 1: Supported cloud providers and regions available for Snowflake (more on regions are listed in the documentation)

When considering what regions on which clouds to use for your primary and secondary source, it’s important to select regions that complement your unique business continuity plan. Consider the following when you choose two regions for your primary and secondary storage:

• Multi-cloud versus single cloud: A multi-cloud strategy helps ensure availability and resiliency if a cloud provider experiences an outage in multiple regions. It can also ensure data portability in the event that an exit plan needs to be in place. If you choose a single cloud, consider taking advantage of functionality such as Azure Regional Pairs. 

• Geographic isolation: Choose two regions that are far enough from each other that a natural disaster, civil unrest, or other regional event in one region won’t affect the other region. If it’s possible in your geographical area, choose a distance between the two of 250–300 miles (or 400–500 km), which is ideal. 

• Data residency: Some data must stay within national borders to comply with regulations. When choosing your regions, consider what data you’re storing and whether any regulations apply. 

• Cost: Consider how you might move consumption to cheaper regions and across cloud providers. With greater data portability built into your data strategy, you’re able to move production systems more easily and can take advantage of differentials in underlying cloud provider rates. 

• Performance: Capitalize on regional footprints to leverage the best cloud provider by region based on its presence, capacity, and services for local teams. The secondary region can be used to help consumers in that region gain performance advantages. 

Once you select which regions work best for your business, it’s easy to set up cross-cloud and cross-region replication and failover/failback within Snowflake. It’s as simple as going to a database and enabling replication and selecting regions. See the demo below.

To learn more about how replication and failover/failback work, see the Snowflake documentation.