Announcement

Collapse
No announcement yet.

Running The Native ZFS Linux Kernel Module, Plus Benchmarks

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Running The Native ZFS Linux Kernel Module, Plus Benchmarks

    Phoronix: Running The Native ZFS Linux Kernel Module, Plus Benchmarks

    In August we delivered the news that Linux was soon to receive a native ZFS Linux kernel module. The Sun (now Oracle) ZFS file-system has long been sought after for Linux, though less now since Btrfs has emerged, but incompatibilities between the CDDL and GPL licenses have barred such support from entering the mainline Linux kernel. There has been ZFS-FUSE to run the ZFS file-system in user-space, but it comes with slow performance. There has also been work by the Lawrence Livermore National Laboratories in porting ZFS to Linux as a native Linux kernel module. This LLNL ZFS work though is incomplete but still progressing due to a US Department of Energy contract. It is though via this work that developers in India at KQ Infotech have made working a Linux kernel module for ZFS. In this article are some new details on KQ Infotech's ZFS kernel module and our results from testing out the ZFS file-system on Linux.

    http://www.phoronix.com/vr.php?view=15474

  • jalyst
    replied
    Originally posted by devsk View Post
    The project is in nascent state but stable. Stability not performance has been the goal so far.

    Performance will improve with time. In my experience, its definitely better than zfs-fuse at this time but not upto BTRFS or ext4.

    Reads scale well in RAIDZ.
    Thanks for clarifying!
    Might have a closer look in a yr or so, after I've played much more with BTRFS etc.

    Leave a comment:


  • devsk
    replied
    Originally posted by jalyst View Post
    Nice find, how does it compare to the latest stable BTRFS all up?
    I'm sure it's much more featureful, but aside from that how does it compare.
    The project is in nascent state but stable. Stability not performance has been the goal so far.

    Performance will improve with time. In my experience, its definitely better than zfs-fuse at this time but not upto BTRFS or ext4.

    Reads scale well in RAIDZ.

    Leave a comment:


  • jalyst
    replied
    Originally posted by devsk View Post
    Any plans to compare the latest zfs-on-linux code from LLNL (as opposed to KQI code you used earlier)? It runs stable on 2.6.38 here.

    http://zfsonlinux.org/
    Nice find, how does it compare to the latest stable BTRFS all up?
    I'm sure it's much more featureful, but aside from that how does it compare.

    Leave a comment:


  • devsk
    replied
    Any plans to compare the latest zfs-on-linux code from LLNL (as opposed to KQI code you used earlier)? It runs stable on 2.6.38 here.

    http://zfsonlinux.org/

    Leave a comment:


  • mm-freebsd
    replied
    Native ZFS on Ubuntu and Fedora Linux

    KQ Infotech has officialy made available the source code for native ZFS linux modules.
    More info at kqstor.com

    You can create and install packages for Ubuntu (Debian) and Fedora Linux using the tutorial at blog.vx.sk.

    Leave a comment:


  • devsk
    replied
    Originally posted by kraftman View Post
    You've agreed with the most famous troll from osnews.com.
    And this post makes you what?

    I would stand up for people when they say something right. Troll or not! Amazing what bias can do to our thought process and responses.

    And then there are people like you. Who label others just for their strong beliefs. Shame on you!

    Leave a comment:


  • kraftman
    replied
    Originally posted by devsk View Post
    what's the configuration? All 48 drives converging onto a single ext4 filesystem? If not, then I think you really don't know what he is talking about. He was talking about ZFS managing large number of disks and scaling performance better than Linux filesystems. And I agree with him.
    You've agreed with the most famous troll from osnews.com.

    Leave a comment:


  • devsk
    replied
    Originally posted by phoronix View Post
    Phoronix: Running The Native ZFS Linux Kernel Module, Plus Benchmarks

    In August we delivered the news that Linux was soon to receive a native ZFS Linux kernel module. The Sun (now Oracle) ZFS file-system has long been sought after for Linux, though less now since Btrfs has emerged, but incompatibilities between the CDDL and GPL licenses have barred such support from entering the mainline Linux kernel. There has been ZFS-FUSE to run the ZFS file-system in user-space, but it comes with slow performance. There has also been work by the Lawrence Livermore National Laboratories in porting ZFS to Linux as a native Linux kernel module. This LLNL ZFS work though is incomplete but still progressing due to a US Department of Energy contract. It is though via this work that developers in India at KQ Infotech have made working a Linux kernel module for ZFS. In this article are some new details on KQ Infotech's ZFS kernel module and our results from testing out the ZFS file-system on Linux.

    http://www.phoronix.com/vr.php?view=15474
    Any plans to repeat the same with latest code on top of 2.6.35.10 kernel? It upgrades to more recent ZFS codebase as well as fixes a lot of issues. In my own experiences, the latest code is stable and very performant, even on a USB drive.

    Leave a comment:


  • gilboa
    replied
    Originally posted by devsk View Post
    what's the configuration? All 48 drives converging onto a single ext4 filesystem? If not, then I think you really don't know what he is talking about. He was talking about ZFS managing large number of disks and scaling performance better than Linux filesystems. And I agree with him.

    We have a Linux server with 68 drives but that doesn't tell anything about how the filesystem is using those 68 drives.
    Hence the reason I said he had little idea what he's talking about - ext4 doesn't handle physical volumes - its not btrfs or ZFS.
    If you have large RAID (Say, above 6-8 drives), you don't use a ext4/LVM/SoftMD combo - you simply buy an expensive SAS RAID controller w/ RAID6 support and huge amounts of memory. (In my case: multiple ext4 partitions over a single LVM PV running on an external FC box)

    - Gilboa

    Leave a comment:

Working...
X