Slashdot Mirror


Oracle Engineer Talks of ZFS File System Possibly Still Being Upstreamed On Linux (phoronix.com)

New submitter fstack writes: Senior software architect Mark Maybee who has been working at Oracle/Sun since '98 says maybe we "could" still see ZFS be a first-class upstream Linux file-system. He spoke at the annual OpenZFS Developer Summit about how Oracle's focus has shifted to the cloud and how they have reduced investment in Solaris. He admits that Linux rules the cloud. Among the Oracle engineer's hopes is that ZFS needs to become a "first class citizen in Linux," and to do so Oracle should port their ZFS code to Oracle Linux and then upstream the file-system to the Linux kernel, which would involve relicensing the ZFS code.

3 of 131 comments (clear)

  1. Re:Mainstream in FreeBSD... by Anonymous Coward · · Score: 1, Interesting

    Holy shit are you serious? Like SERIOUS? OMG why don't we all switch to BSD! Everyone stop! I know Linux is *everywhere* but BSD has ZFS! Did you guyz know this????

  2. Re:Having it NOT be in upstream is more flexible by JBMcB · · Score: 5, Interesting

    Funny, I thought ZFS was very mature by now.

    It's very mature, on Solaris. Linux has a different ABI to the storage layer, and different requirements on how filesystems are supposed to behave. So it's not so much a port as a re-implementation.

    --
    My Other Computer Is A Data General Nova III.
  3. Re:Not the best fit since it's schizophrenic by Anonymous Coward · · Score: 2, Interesting

    ZFS mashes all those different things into one big blob. That's not really how Linux is designed.

    That's because Linux isn't designed, it's grown organically in a hodgepodge fashion. Some people think this is a good thing. Others do not.

    A weblog post by Jeff Bonwich, one of the creators of ZFS, from ten years ago**:

    Andrew Morton has famously called ZFS a "rampant layering violation" because it combines the functionality of a filesystem, volume manager, and RAID controller. I suppose it depends what the meaning of the word violate is. While designing ZFS we observed that the standard layering of the storage stack induces a surprising amount of unnecessary complexity and duplicated logic. We found that by refactoring the problem a bit -- that is, changing where the boundaries are between layers -- we could make the whole thing much simpler.

    https://blogs.oracle.com/bonwick/rampant-layering-violation

    He gives a reasonable answer as to why glomming all that together has its advantages. Good intro slide deck:

    https://wiki.illumos.org/download/attachments/1146951/zfs_last.pdf

    Note that "ZFS" is actually made of of three layers: the SPA (which talks to disks), the DMU (which takes objects and breaks up into the RAID stripes to send them to the SPA), the ZPL (ZFS POSIX layer, which is your Unix-y file system).

    You can actually link directly to the DMU (which has a userland library) and treat "ZFS" as an pure object store without POSIX semantics. You could also take another file system (ext3/4, UFS, XFS) and plug it into the DMU as well, and treat the lower layers as a replacement to LVM.

    ** Ten years? Holy shit! I remember reading that shortly after it was posted.