No announcement yet.

Btrfs Gets Big Changes, Features In Linux 3.14 Kernel

  • Filter
  • Time
  • Show
Clear All
new posts

  • #51
    Originally posted by mercutio View Post
    true, i'd like to see 3 disk raid 10 support, but i haven't seen any work started on that. and i think lz4 support would benefit more users. i'd ilke to see btrfs with lz4 compression start appearing on cellphones for instance.
    Jolla phone uses BTRFS, not sure which compression though.


    • #52
      Originally posted by erendorn View Post
      What percentage of data on your system is an uncompressed book? I'm not taking any sides, I have no idea if it is or not worthwhile, but this benchmark is not that usefull.
      Do we have boot times? boot time of a virtual machine stored on a btrfs fs? time to backup a system disk? a user data disk, with steam games (maybe compressible) and video/pictures (incompressible)? Compilation times without cache, or time to clone/update a big git repo?

      I don't doubt that LZ4 is faster, the question is, is it the bottleneck, and what gains can we expect in real life.
      i did it with /usr/bin/clang too much was my biggest executable in /usr/bin and clang gave better compression. we don't have any real times, because btrfs isn't in mainline kernel yet and grub doesn't support it yet.

      compilation times seem to be close enough to not care about between different filesystems. (ext4, btrfs with lzo compression, zfs with lz4 compression)

      boot times don't really matter too. what seems to matter, is how much it slows your system down when doing lots of io, where btrfs either has bugs or compression overheads.
      Last edited by mercutio; 02-01-2014, 05:19 PM.


      • #53
        Originally posted by mercutio View Post
        we don't have any real times, because btrfs isn't in mainline kernel yet and grub doesn't support it yet.
        Uh, what?


        • #54
          Originally posted by GreatEmerald View Post
          Uh, what?
          err btrfs with lz4 i meant.


          • #55
            Originally posted by Prescience500 View Post
            Is there or will there be an effort to make ensure that BTRFS is as fast as or faster than EXT4? I know BTRFS is all about features rather than performance, but the average home user doesn't need all of those advanced features. For me, faster makes a less painful time redoing my operating system and transfering all of my files every 6 months.
            This perspective is coming from someone using ZFS but all of the features of these file systems are definitely applicable to home users. (I'm pretty sure btrfs has most of these same features but syntax is different)

            On disk compression results in improved performance and less space used.

            Snapshots mean you could snapshot your system before you let your mom on the computer and after she pollutes your browser history with pinterest, aol email and browser toolbars, you could restore back to the snapshot.

            You can also use snapshots to restore computer after a hardware failure. I needed to rebuild the zfs pool on my desktop a while back to switch drive layout and to back it up I did a zfs send/recv to my nas, rebuilt the pool and then zfs send/recv back to the desktop to restore my exact system.

            Then checksums ensures that everything is perfectly preserved. In some cases it can give you an early warning that a drive is dying and in others, it can prevent movies/pictures of loved ones from corrupting.

            Then, folder mount points on ZFS also make for extremely easy migration from distro to distro. As an example of this, You can install a secondary OS alongside another and switch from one to the other while maintaining all files from the old install without juggling partitions. Just leave entire drive as ZFS:

            zfs create pool/OS -o mountpoint=none
            zfs create pool/OS/Arch -o mountpoint=/
            zfs set mountpoint=/pool pool
            zpool set bootfs=pool/OS/Arch pool

            Then lets say you didn't like arch and wanted to use gentoo instead. All you would have to do is:

            zfs set mountpoint=/backup pool/OS/Arch
            zfs create pool/OS/Gentoo -o mountpoint=/
            zpool set bootfs=pool/OS/Gentoo pool

            This would switch your install to Gentoo while keeping your entire arch install intact. You could switch back if needed or copy the needed files over and then delete it. You could also do this with your home directory to just directly mount your arch home directory into your gentoo install in the exact same spot without needing to copy files over.

            Also, in regards to your performance concern, a ZFS raid on more than 2 disks (e.g. a raid 5,6,10) outperforms ext4 + mdadm in virtually all cases, zfs also outperforms ext4 + mdadm on a raid 1/0 in multiuser configs (like a server serving files to 5+ clients at once). And that is in just a basic, no frills configuration. With ZFS, you can turn on compression and add SSD caches to make that even faster.