this post was submitted on 14 May 2024
186 points (98.4% liked)
PC Gaming
8597 readers
644 users here now
For PC gaming news and discussion. PCGamingWiki
Rules:
- Be Respectful.
- No Spam or Porn.
- No Advertising.
- No Memes.
- No Tech Support.
- No questions about buying/building computers.
- No game suggestions, friend requests, surveys, or begging.
- No Let's Plays, streams, highlight reels/montages, random videos or shorts.
- No off-topic posts/comments, within reason.
- Use the original source, no clickbait titles, no duplicates. (Submissions should be from the original source if possible, unless from paywalled or non-english sources. If the title is clickbait or lacks context you may lightly edit the title.)
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
My solution is RAIDZ5 and storing the backup on LTO6 tape with parity/erasure code. I think the fact that scrub times take 24 hours even on 16TB drives is already over the safety margin. If a drive failure happens, the first thing I'll do to run a manual diff backup which should take a fraction of the time and then run the ZFS resilver.
I'm beginning to see why SSD RAID is being considered now. My guess for HDDs in enterprise is that a RAID 15 (I made this up) would be considered. What I mean is data is stored on two identical servers each running RAID5 or 6. Off the shelf solutions like Gluster exist and that seems to be gaining traction at least according to Linus Tech Tips.
SSD RAID is actually very common outside of home use! And yeah, clustered filesystems help overcome many of these limitations, but tend to be extremely demanding (expensive hardware for comparable performance). Network almost immediately becomes the bottleneck. Even forgetting about latency and other network efficiency concerns, 100 Gbps isn't that fast when you have individual devices approaching 16 Gbps.