AlloyDB protects your data through two distinct backup mechanisms, and the Professional Cloud Database Engineer exam expects you to know which one applies in a given scenario. The first is continuous backup and recovery, which is the default protection method and is enabled automatically on every cluster. The second is discrete backups, which are standalone, complete copies of the database taken at a specific moment. Both produce a usable recovery path, but they differ in how they capture data, how precisely you can restore, and what they are typically used for.
Continuous backup and recovery is enabled by default on all AlloyDB clusters, so you have a safety net from the moment you start using the service without configuring anything. Rather than capturing the database at discrete intervals, this method records changes as they happen. That continuous capture is what gives it a recovery point objective, or RPO, of zero. Because every change is retained, you do not lose committed data in the event of a failure.
The result is a point-in-time restore with microsecond precision. You pick a specific moment within a retention window and AlloyDB creates a new cluster from the recent state of an existing cluster as it was at that point. The retention window is configurable from 1 to 35 days, and the default is 14 days. The restore lands within the same region as the source cluster, which makes this the mechanism to reach for when you want to clone a cluster or recover from data corruption or an unwanted change by rolling back to just before it occurred.
Discrete backups work differently. Each one is a complete copy of the database taken at a specific moment, and it stands on its own as a fixed point in time. You use a discrete backup to rebuild a cluster from that captured state when you need it. AlloyDB supports creating these backups either on a schedule or manually, which gives you flexibility for cases like long-term archiving or taking a snapshot right before a maintenance operation.
The distinction the exam tends to lean on is precision versus discreteness. Continuous backup and recovery lets you restore to any microsecond inside the retention window, so it is the right answer when the requirement is point-in-time recovery or an RPO of zero. Discrete backups give you specific, named copies at the moments you chose to take them, which suits scheduled snapshots and retained archive points rather than arbitrary-moment recovery.
For the Professional Cloud Database Engineer exam, anchor on a few facts. Continuous backup and recovery is on by default, achieves RPO equals zero, restores with microsecond precision within a 1 to 35 day window that defaults to 14 days, and produces a new cluster in the same region. Discrete backups are complete standalone copies, created on a schedule or manually, used to rebuild a cluster. When a scenario calls for recovering to an exact point in time or losing no committed data, that points to continuous backup and recovery. When it describes periodic snapshots or archival copies, that points to discrete backups.
Our Professional Cloud Database Engineer course covers AlloyDB backups alongside point-in-time recovery and cross-region disaster recovery, with practice questions that drill these distinctions.
