Not sure if any of that is helpful for your case but I recommend trying something if you’ve got spare hardware, and see how it goes on dummy data, then blow it away try something else.

This is good advice, thanks! Pretty much what I’m doing right now. Already tried it with IPFS, and found that it didn’t meet my needs. Currently setting up a tahoe-lafs grid to see how it works. Will try out ceph after this.

How is ceph working out for you btw? I’m looking into distributed storage solutions rn. My usecase is to have a single unified filesystem/index, but to store the contents of the files on different machines, possibly with redundancy. In particular, I want to be able to upload some files to the cluster and be able to see them (the directory structure and filenames) even when the underlying machine storing their content goes offline. Is that a valid usecase for ceph?

Yep. Intel atom D525