Disaster Recovery

Disaster recovery protects your data against the loss of an entire region. High Availability and Backups both keep your data within a single region, so neither survives a regional outage. To prepare for one, you keep a copy of your data in a separate region.

Export to another region

VeloDB Cloud bases disaster recovery on cross-region export. You export the databases that you need to object storage in another region. The export RPO is the export cycle you set plus the time each export takes to complete, because the recovery point is your last completed export, so schedule exports to match your target RPO. After a regional outage, you create a warehouse in the second region, import the exported data, and redirect your applications to it. Both the export and the import take longer as your data grows, so together they determine your RTO. See Export Overview for the available methods and commands.

Dual-write to two regions

Export-based recovery loses any data written since the last export and takes time to import, so its RPO and RTO are bounded by how often you export and how much data you move. When you need a smaller RPO and RTO, run a warehouse in each of two regions and have your application write to both. If one region fails, the other already holds a current copy and continues to serve traffic, which brings both RPO and RTO close to zero. This approach costs more, because you operate and load two warehouses instead of one.

See also

• Export Overview describes the export methods and commands.
• High Availability explains how multi-zone deployment survives the loss of a single availability zone.
• Reliability compares disaster recovery with high availability and backups.