Backups overview

Backups give you the ability to restore data to a point in the past, as well as to export and analyze data. Backups are useful for the following scenarios:

  • Disaster recovery: use backups as part of a disaster recovery plan. In the event of a disaster, you can restore data to a new cluster in Memorystore for Redis Cluster. Backup schedules provide more robust protection than ad hoc exports.

  • Data migration: migrate data between different clusters to consolidate resources or move data to a different region.

  • Data sharing: share data across different teams or applications to enable collaboration, offline data analytics, and data exchange.

  • Compliance: create periodic backups of cache data to meet compliance requirements.

You can't import Memorystore for Redis Cluster backups into Memorystore for Redis instances because they contain multiple RDB files.

Backups and backup collections

A backup is a point-in-time snapshot of your data. You can organize backups into backup collections. A backup collection includes all backups for a cluster. When you delete a cluster, Memorystore for Redis Cluster doesn't delete the corresponding backup collection automatically.

You can set a time to live (TTL) on a backup by configuring a retention period. After a backup exceeds its TTL, Memorystore for Redis Cluster deletes it automatically.

To recover data from a backup, you must create a cluster and restore the data to it. You can't import backup data into an existing cluster.

Automated and on-demand backups

You can initiate backups in the following ways:

  • Automated backups (Recommended): configure a backup schedule to initiate periodic backups for a cluster.

  • On-demand backups: initiate a backup operation that begins immediately.

To learn how to set an automated backup schedule or to initiate a backup, see Manage backups.

Pricing

You're charged storage costs for each backup based on its size. For each backup, you're charged a minimum of 24 hours. For more information, see Memorystore for Redis Cluster pricing.

What's next