Between the ZFS-like features and the advertised "novel copy-on-write metadata" scheme, I would not be surprised if it was partially based on DragonFly's HAMMER/HAMMER2 filesystem.
If you poke around in the the APFS kernel extension, it's not a very big binary, given the feature set. (It's a 550K extension, compared to the 2.5MB zfs.ko on FreeBSD.) I haven't disassembled it yet, but I'd wager that APFS is layered atop Core Storage btrees/CoW. Since that code's been shipping since 2011, maybe APFS stabilizes faster than a true greenfield filesystem?
(That does leave me wondering how interesting an open-source APFS would be without an open-source Core Storage.)
> It's a 550K extension, compared to the 2.5MB zfs.ko on FreeBSD
hammer.ko is 494KB in the latest DFBSD, for those wondering the obvious.
I was thinking the exact same thing!
It would be fantastic if they actively helped with HAMMER2 development.
It would be a strange kind of NIH that excluded ZFS but welcomed HAMMER...
different licensing and HAMMER demands significantly less memory
Re:Apple File System
by AceJohnny2 on Monday June 13, 2016 @08:52PM
(#11899012)
Licensing.
Yes. I was a heavy user of ZFS on Mac OS X back then. Was on the developer mailing list for that implentation, submitting bugs and talking to Apple developers working like mad to ship it.
Then Steve Jobs' buddy Larry bought Sun, and licensing of ZFS became basically impossible to sort out on time. So they dropped it.
Was that 2007 or 2009?
2009. I remember ZFS was coming in Snow Leopard. Then it wasn't
/edit
https://web.archive.org/web/20080721031014/http://www.apple....
The Sun acquisition was announced in 2009 and closed in 2010. It may have been related, but I have my own suspicions about about what caused Apple to jettison their port.
Given the mountain of legal shit Google has been subjected to over Java, the last thing Apple wanted was to have to answer to Larry's lawyers. ZFS, awesome as it is, has been poisoned.
It's like no one here remembers the NetApp lawsuit.
Re:Apple File System
by rafaelferreira on Monday June 13, 2016 @10:02PM
(#11899262)
Do tell.
The thing that killed ZFS on the Mac was the NetApp lawsuit over it against Sun.
It's impossible to know how long they have been working on it. As I recall, Swift had been in development for 3+ years before it saw the light of day.
Yeah, I bet it's been in development for a while. People have been seriously bitching about HFS+ for at least a decade, and with good reason. Apple has been pretty much silent about it apart from briefly testing ZFS and I'd wager a guess that they started working on it shortly after they abandoned that, while they continued also bolting more stuff onto HFS+ in the meantime.
Unless I'm misreading, the page seems to indicate that only external volumes can be formatted with APFS. My guess is that they'll have limitations and HFS-defaults for several years, at least on macOS, as the file system matures.
Either way, what file system didn't take years to get right? There are so many possible edge cases with file systems that it not only takes a long time to sort them out, but an amazing community and/or luck to reproduce deterministically to fix them.
File systems take A LONG time to stabilize, and you want them stabilized well, because if there is a bug in httpd, you just restart it, if there is a bug in filesystem you loose your data. See BTRFS. Started in 2007 and it is still not stable enough to be supported in RHEL (and for good reasons). So, of course, Apple engineers are way better than the rest of the world and they have magical pixie dust which makes bugs disappear, so they will fix bugs way faster than anybody else. Not.
It's good to see Apple catching up, no matter where it's coming from. You mention ZFS, but per-file encryption (EFS) and snapshots (VSS) in particular stood out to me as features NTFS has had for a decade.
Anyone know who works on APFS? If I were Apple I would have certainly picked up some of the ZFS core team, curious if any of them are currently at Apple.
Could speed up their time to market with such seasoned hands on board.
One of them worked on the 'Be File System'... that's interesting
https://en.wikipedia.org/wiki/Dominic_Giampaolo
It seems ZFS was a no-go because of its license. But why does Apple develop its APFS rather than port the open source BTRFS with all its features? NIH syndrome?
On-disk consistency ("crash protection"), snapshots, encryption, and transactional interfaces ("atomic safe-save") will no doubt be incredibly valuable. I don't think though though that APFS will dramatically improve upon the time it took ZFS to mature from a first product to world-class storage.
Some commenters have opined that (despite Apple distributing ZFS for Mac OS X at WWDC nearly a decade ago) that ZFS would never be appropriate for the desktop, phone, or watch. True ZFS was designed for servers and storage servers, but I don't think there's anything that makes it innately untenable in those environments--even its default, but not essential, use of lots of RAM.
Who knows... maybe Apple have spent the decade since killing their internal ZFS port taking this new filesystem though the paces. Its level of completeness though would suggest otherwise.