Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

The docs state:

"On systems with 1,000s of ZFS file systems, provision 1 GB of extra memory for every 10,000 mounted file systems including snapshots."

What does that mean - how do these people use ZFS? Are they using ZFS inside per-user mounted files? Or does it refer to really huge datacenters??

I always thought ZFS advocates for a single file-system, a giant one. Why do the docs talk about people mounting 1000s??

share|improve this question
I'm struggling to find an answer to this question, we're a Q&A site, not a discussion site - could you edit your question to make it more easily answerable please. If it's just "Do people really mount tens of thousands of ZFS filesystems" then I'm not sure what benefit the question would have other than out of curiosity. – Chopper3 Dec 13 '11 at 18:35
You actually catched me misusing serverfault - agreed. – isync Dec 13 '11 at 19:49
up vote 4 down vote accepted

A potential use-case for ZFS that would see thousands of filesystem mounts would be an organization that uses ZFS for NFS-mounted home directories. Imagine a large company or small university with zpools dedicated to home directory exports. Each user could have a dedicated ZFS filesystem with separate quotas and parameters. If rolling snapshots are implemented, it would be possible for users to see previous versions of their files by descending into the ~/.zfs/snapshot directory. Self-service restores!

share|improve this answer
That's what I oversaw: I was so consumed by the perspective of having a central /home dir and populating userdirs in there that I oversaw that it might be nice to separate/ lock users into their own filesystem (!), while still being able to let ZFS deduplicate, compress etc. one level below, technically on block level (right?). I think it came from me being very shocked the other day, when I saw people backing up solid 90GB VHDs containing one nearly empty fs each. – isync Dec 13 '11 at 19:51
Be aware that ZFS de-duplication is not really supported in production on ZFS appliances. – pfo Dec 13 '11 at 19:57
And I am on Linux... No dedup anytime soon. Was dreaming... I guess. – isync Dec 13 '11 at 21:32
right, so given this situation given enough disk can zol handle 10s of thousands of filesystems? it is very memory intensive? – Stu Sep 19 '15 at 19:25
@Stu Sure it can. – ewwhite Sep 19 '15 at 19:28

You are missing some of the basic concepts of ZFS. ZFS is meant to be one large pool of disks. After you have established the pool, you are suposed to put file systems on the disks of the pool. All of these file systems then benefit from the pool's underlying RAID layout and other IO characteristics (ZIL et al). Seeing 1000s of file systems reside on a pool is a possible scenario on larger systems.

share|improve this answer
OK. For me ZFS was mostly a file-system. My mistake: I didn't separate that ZFS does pooling and file-system, as I would use it to pool and build one big fs inside. – isync Dec 13 '11 at 19:40

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.