With data deduplication, you can save a significant amount of cloud storage space and reduce operational costs. Cloudback offers a simple but efficient technique of data deduplication.
How deduplication works
Our backup archives are deterministic. If nothing has changed in a GitHub repository, we create the same archive before encrypting it. Once the archive is encrypted, the AES encryption makes it non-deterministic.
Cloudback compares the checksum of a new deterministic backup with a previous deterministic backup before encrypting it. If the archives match, Cloudback doesn’t upload a new archive to storage, but instead stores a link to the previous archive. The retention policy is extended for the previous archive, it’s not deleted until there are valid linked backups.
Deduplication occurs when certain conditions are met:
- Deduplication is enabled for cloud storage in the
New storageor the
- The previous backup of a repository is stored in the same cloud storage
- The checksum of a previous backup matches the checksum of a new backup
How to enable deduplication
The feature is released on 06 May 2022, please note that:
- Deduplication is disabled by default for all customer managed storages created before 06 May 2022; you must enable it manually if you want to use this feature.
- Deduplication is enabled by default for all customer managed storages created after 06 May 2022; you must manually disable it if you don’t want to use this feature.
Data deduplication is a customer managed storage setting. It’s configured separately for each storage. Use an additional
Deduplication Type combo box in the
New Storage and
Edit Storage dialogs to control the behavior of data deduplication:
How to check if a backup is deduplicated
In the dashboard, on the repository card, you’ll find the
Backup Information tooltip available for all backups. Within the tooltip there’s an additional label
Deduplication. There it says
Yes if the backup has been deduplicated. See the screenshot below: