First off, I personally wouldn’t touch WD drives with a 10′ pole. They are too expensive to recover in event of catastrophic failure… often 2-3x the cost of data recovery vs. Seagate due to the way their spindles are affixed. So next point is that RAID != Backup, and vice versa. RAID reduces downtime. Backup ensures the integrity of your backup. If a single drive failed, you wouldn’t want to have to download your whole backup to recover. That’d cost a lot, and would be a huge waste of time. So we RAID our storage servers to avoid any downtime (a drive fails, you replace it and all is still running, without having to resort to our backups). But then, if a surge hits taking out all your drives, or a flood wipes out your server, THAT is when your backup comes into play. Personally, I have an unRAID server with 4 TB space on-site, and a 2TB external hard drive connected to a Pogoplug off-site. My unRAID server backs up to it (rsync) nightly. It’s the best protection I can get for next-to-nothing. I wouldn’t do AWS-based backup due to the same reason it’s impractical to you: the cost. Just find a friend or family member that’ll host a Pogoplug in a closet somewhere across town, and automate everything.