|
|
|
|
|
by frontierkodiak
1519 days ago
|
|
I've been very happy with self-hosting a Seafile instance to serve as a personal cloud for myself and a small academic ML team. I built a small NAS, Proxmox runs on baremetal, on top of which there's an ubuntu VM running a Seafile docker instance. Fantastic performance, so so much more efficient for syncing large libraries b/w machines. Owncloud/Nextcloud was a pig for my purposes, it seemed unable to sync libraries w/ large number of files (2million+). IIRC Seafile sync client is C under the hood, and is much more performant than any other sync client I've tried. OneDrive was completely unable to upload the amt of data that I needed, and I was tired of manually splitting my datasets into chunks, only to have to stitch them back together on the shitty WebUI. Feel free to hit me with any questions if anyone is thinking of setting up a Seafile instance. I've found their 'community edition' free release to be more than sufficient for our team. |
|