Page 1 of 3 123 LastLast
Results 1 to 10 of 25

Thread: Benchmarks Of The Official KQ ZFS Linux Module

  1. #1
    Join Date
    Jan 2007
    Posts
    15,088

    Default Benchmarks Of The Official KQ ZFS Linux Module

    Phoronix: Benchmarks Of The Official KQ ZFS Linux Module

    Last summer we delivered the news that a native ZFS file-system implementation for Linux was coming by an Indian company known as KQ Infotech where they leveraged the Lawrence Livermore National Laboratories ZFS Linux code, finished it off in some areas, and took care of the POSIX support. This ZFS Linux module was eventually released to a group of beta testers -- us included -- and we ran some ZFS Linux benchmarks back in November using the latest beta code. Since that point, however, KQ Infotech has made their ZFS Linux port publicly available and earlier this month they declared this work as stable via its general availability release. We have decided to benchmark this latest ZFS Linux code to see where the performance now stands against the EXT4, Btrfs, and XFS file-systems.

    http://www.phoronix.com/vr.php?view=15656

  2. #2
    Join Date
    Mar 2008
    Posts
    575

    Default

    Interesting... ZFS support Linux is simply a good thing, exactly like ntfs-3g.

  3. #3
    Join Date
    Dec 2007
    Location
    Edinburgh, Scotland
    Posts
    589

    Default

    It would be interesting to see the comparisons on an enterprise / business machine with production work loads. I'm guessing that's where ZFS is targeted at.

    It's really impressive what the KQ folks have done

  4. #4
    Join Date
    Jul 2008
    Posts
    20

    Default

    > Besides the pure file-system performance, each file-system carries its own set of pros and cons in terms of other features and capabilities and can be greatly tuned to enhance the disk performance -- i.e. with Btrfs there is disk compression support and now the space cache option as well.


    To provide some balance to that statement: ZFS also has compression support, and has a deduplication option as well. AFAIK there is no equivalent to the "space cache" option. I can't help but mention that ZFS supports self-healing if using > 1 disk.


    btw: "i.e." means "that is", or "in other words". "E.g." would have been more appropriate when giving examples of features and capabilities.

  5. #5
    Join Date
    Dec 2009
    Posts
    73

    Default

    Nice perform by ZFS.

    I'm surprised btrfs outperformed ext4 in most of the test, in contrast to previous benchmarks.

    is it a good moment now to move to this fs from ext4 on an SSD disk?

  6. #6
    Join Date
    Nov 2008
    Posts
    418

    Default

    Fireburn,
    Yes, ZFS is targeted to large Enterprise and many many discs. In a benchmark, BTRFS vs ZFS, BTRFS stopped gaining performance as they added more and more disks. ZFS just gained more and more performance. ZFS scales well. BTRFS does not - it is targeted for small systems with single disk, or a few disks.

    mmmmbop,
    ZFS provides self healing with one single disk, if you specify "copies=2". This means ZFS will store data blocks twice, halving the disk capacity. You can also specify "copies=3".

    Viper_Scull,
    I would not move to BTRFS from ext4, as BTRFS is broken by design, says Edward Shiskin, a RedHat developer:
    http://en.wikipedia.org/wiki/Btrfs
    Find his name on that page, which points to links to the Linux mail list discussions where they discuss if BTRFS is broken or not.

  7. #7
    Join Date
    Jul 2008
    Location
    Greece
    Posts
    3,798

    Default

    Quote Originally Posted by kebabbert View Post
    Viper_Scull,
    I would not move to BTRFS from ext4, as BTRFS is broken by design, says Edward Shiskin, a RedHat developer:
    http://en.wikipedia.org/wiki/Btrfs
    Find his name on that page, which points to links to the Linux mail list discussions where they discuss if BTRFS is broken or not.
    I've just read the LKML thread. I don't see how it's broken. The devs replied to Edward explaining why his concerns don't affect BTRFS. (And there are no further replies from Edward after that.)

  8. #8
    Join Date
    Oct 2008
    Posts
    14

    Default reiser4

    Would it be that hard to also include reiser4 in these sorts of tests?

  9. #9
    Join Date
    Sep 2006
    Posts
    714

    Default

    Quote Originally Posted by kebabbert View Post
    Fireburn,
    Yes, ZFS is targeted to large Enterprise and many many discs. In a benchmark, BTRFS vs ZFS, BTRFS stopped gaining performance as they added more and more disks. ZFS just gained more and more performance. ZFS scales well.
    That is because BTRFS is programmed in C code, while ZFS is written in magic pixie dust.

    BTRFS does not - it is targeted for small systems with single disk, or a few disks.
    For god sakes keep this factoid to yourself, otherwise the experienced enterprise file system programmers from IBM, Redhat, and Oracle might think that there is no point continuing to develop BTRFS and I want a nice file system for my netbook.

  10. #10
    Join Date
    Oct 2008
    Posts
    3,173

    Default

    Quote Originally Posted by RealNC View Post
    I've just read the LKML thread. I don't see how it's broken. The devs replied to Edward explaining why his concerns don't affect BTRFS. (And there are no further replies from Edward after that.)
    Yes, that's already been explained to kebabbert several times. He keeps spamming the same links over and over again, though.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •