Managing compute backups
A backup, or recovery point (these terms are used interchangeably), can be a copy either of a compute volume or only of data changes that is made at a specified time. With the backup service, you can create backups automatically by using backup plans or initiate backups manually. Backup plans define what data to back up, how frequently to create backups, and how long to keep them.
The backup service also allows you to restore virtual machines and volumes by creating new instances from backups.
The following backup types are supported:
- Full. A full backup contains a copy of an entire compute volume. It is self-sufficient, meaning that you can restore data from just one such a backup. Full backups are time-consuming and take up a large amount of storage space.
- Incremental. An incremental backup only copies changes to the data since the latest backup, regardless of its type. Incremental backups are usually faster and require less storage space. However, restoring data from incremental backups is more complex, as it requires the full backup chain consisting of the first full backup and all subsequent incremental backups.
Limitations
- You need a license to be able to create and manage incremental backups.
- You cannot back up the same volumes by using both the integrated backup service and an external backup solution.
Prerequisites
- The compute cluster is created, as described in Creating the compute cluster.
- The backup service is enabled in the command-line interface, as described in Provisioning the backup service.