Announcement

Collapse
No announcement yet.

Btrfs Mount Option Benchmarks With The Linux 4.9 Kernel

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

  • Btrfs Mount Option Benchmarks With The Linux 4.9 Kernel

    Phoronix: Btrfs Mount Option Benchmarks With The Linux 4.9 Kernel

    Following last week's Btrfs / EXT4 / XFS / F2FS benchmarks using the Linux 4.6 through 4.9 kernels, some requests came in for doing some fresh Btrfs mount option comparison benchmarks. Thus for your viewing pleasure ahead of the holidays are some Btrfs mount option tests with the Linux 4.9 kernel.

    http://www.phoronix.com/vr.php?view=23894

  • #2
    IIRC, nodatacow still allows snapshots, copying the file only once after it is changed for the first time after a snapshot. Which other features suffer from nodatacow? Data deduplication?

    Comment


    • #3
      Originally posted by treba View Post
      IIRC, nodatacow still allows snapshots
      It doesn't. You loose *all* the features, including checksumming.
      ## VGA ##
      AMD: X1950XTX, HD3870, HD5870
      Intel: GMA45, HD3000 (Core i5 2500K)

      Comment


      • #4
        Seems very strange that discard would show better performance.

        Comment


        • #5
          Originally posted by Sloth View Post
          Seems very strange that discard would show better performance.
          Yes, I was wondering about that too... Aren't Samsung 8** buggy and blacklisted in the kernel?

          Comment


          • #6
            Originally posted by treba View Post
            IIRC, nodatacow still allows snapshots, copying the file only once after it is changed for the first time after a snapshot. Which other features suffer from nodatacow? Data deduplication?
            Well it's kind of pointless IMHO, COW and especially data COW is one of BTRFS' most attractive features. The only case where I disable it is for VM images (obviously).

            Comment


            • #7
              Michael: Typo:

              Originally posted by phoronix View Post
              while discard alsp helped the results.

              Comment


              • #8
                Originally posted by tildearrow View Post
                Michael: Typo:
                Fixed, thanks.
                Michael Larabel
                http://www.michaellarabel.com/

                Comment


                • #9
                  Originally posted by Elinvention View Post

                  Yes, I was wondering about that too... Aren't Samsung 8** buggy and blacklisted in the kernel?
                  Same here.

                  Comment


                  • #10
                    And still no official lz4 support.

                    Comment

                    Working...
                    X