Redundancy & hot-standby. Each AWS SQL deployment includes a mirror replica setup as a hot-standby with data synchronized up to the speed of shipping data increments between AWS data centers. The AWS SQL service is set up for auto-failover allowing a seamless takeover of the hot-backup in case the master instance fails.
Additionally, the mirrored replica resides in a separate AWS availability zone, and availability zones are physically separate in AWS’s data center, enabling the continuity of StackHawk services even in the unlikely event that one of AWS data centers becomes unavailable.
Snapshots & point-in-time backups. Full database snapshots are taken daily and stored for a minimum of 90 days. Daily backups provide a last resort recovery in case of massive data corruption or loss.