What is truly wrong with individual drives representing individual volumes with manual redundancy only on data needing it?
For instance, let’s say I have a 20TB, a 12TB, and an 8TB drive for a 4-bay NAS, then most RAID setups don’t make sense. Additionally, let’s say I really only have <8TB of data needing redundancy too… So, in this case, is there a real reason to not setup as three individual drives, with manual redundancy of necessary data; like a nightly rsync, etc.?
I get why RAID 5, say, is the most recommended, and also understand that RAID isn’t a true, completely safe backup, nor is the setup described above.
Just curious for perspectives from longer-time datahoarders. I’m a newbie to multi-drive NAS hardware.
(Also, is the scenario I described above even possible on, say, QNap or Synology; or is it only RAID or JBOD?)
Welcome to a great conclusion. Key there is knowing what you need for what data and it sounds like you do.
For my data, I have kept them on individual drives with redundancy (copies and/plus pars) for some data and left some data by themself.
In the past, I have shift between individual drives and raid. These days, you have more options including object stores with cloud, rclone and minio.
These days for me, it’s between ability to move data around system fast, portability of data (fast) via ssd, and immediate data access. Life is short to spent it waiting for compute and processing cycle. Too many old guys RIP these days makes me reprioritize 😏