Re:Ceph: A Journey to 1 TiB/s
by speedgoose on Friday January 19, 2024 @05:14PM
(#39062032)
Also garage. https://garagehq.deuxfleurs.fr/
Re:Ceph: A Journey to 1 TiB/s
by BlackLotus89 on Saturday January 20, 2024 @04:40PM
(#39072656)
Garage seems to only to duplication https://garagehq.deuxfleurs.fr/documentation/design/goals/
> Storage optimizations: erasure coding or any other coding technique both increase the difficulty of placing data and synchronizing; we limit ourselves to duplication.
This is probably a nogo for most use cases where you work with large datasets....
Re:Ceph: A Journey to 1 TiB/s
by speedgoose on Sunday January 21, 2024 @01:20AM
(#39076110)
Yes I mentioned it in the context of homelabs, it’s not an alternative to Ceph or MinIO for large installations.
But the simplicity is appreciated. You can recover all the data using rsync.
Re:Ceph: A Journey to 1 TiB/s
by plagiarist on Saturday January 20, 2024 @12:36PM
(#39070099)
Minio doesn't make any sense to me in a homelab. Unless I'm reading it wrong it sounds like a giant pain to add more capacity while it is already in use. There's basically no situation where I'm more likely to add capacity over time than a homelab.
Re:Ceph: A Journey to 1 TiB/s
by reactordev on Saturday January 20, 2024 @01:24PM
(#39070642)
You get a new nas (minio server pool) and you plug it into your home lab (site replication) and now it's part of the distributed minio storage layer (k8s are happy). How is that hard? It's the same basic thing for Ceph or any distributed JBOD mass storage engine. Minio has some funkiness with how you add more storage but it's totally capable of doing it while in use. Everything is atomic.
[1] https://min.io/