What is backup method of PBS?

[Mucip]

Hi,
I want to know backup type of PBS. I have backup of 2 VM (250GiB + 150Gib) and I have 21 backup sets. This mean (250+150)*21 = 8.4TB
I have 3x6TB disk (Size:11.84TB) in ZFS and I see on Dash board %2.82 used!
I couldn't understand. This is perfect compression or what?

If it backup incrementally? But how?

Regards,
Mucip

 

[Chris]

Hi,

Hi,
I want to know backup type of PBS. I have backup of 2 VM (250GiB + 150Gib) and I have 21 backup sets. This mean (250+150)*21 = 8.4TB
I have 3x6TB disk (Size:11.84TB) in ZFS and I see on Dash board %2.82 used!
I couldn't understand. This is perfect compression or what?

PBS uses a content addressable storage to store (compressed) data chunks, thereby efficiently de-duplicating backup snapshot data. A good starting point is to read https://pbs.proxmox.com/docs/technical-overview.html#technical-overview

If it backup incrementally? But how?

No, each backup snapshot is fully self contained, by only referencing those data chunks it is composed of. Identical data chunks will however be shared over a lot of backup snapshots, the datastore being the logical boundary. The client avoids to re-upload chunks already known to the server, a backup snapshot is incremental in that sense, see https://pbs.proxmox.com/docs/faq.html#is-the-backup-incremental-deduplicated-full

If you perform a garbage collection, the summary will show you the logical size and the on disk size, together with the deduplication ratio for your datastore.

Regards,
Mucip

Hope this helps to clear things up!

 

[Mucip]

Dear Chris,
Very valuable and sufficient information. Thanks.

Regards,
Mucip