Announcement

Collapse
No announcement yet.

Btrfs, EXT4 & ZFS On A Solid-State Drive

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

  • Btrfs, EXT4 & ZFS On A Solid-State Drive

    Phoronix: Btrfs, EXT4 & ZFS On A Solid-State Drive

    With the benchmarks recently looking at the performance of ZFS on FreeBSD versus EXT4/Btrfs on Linux having generated much interest and a very long discussion, this morning we are back with more benchmarks when running ZFS on FreeBSD/PC-BSD 8.1 and Btrfs and EXT4 on an Ubuntu Linux 10.10 snapshot with the most recent kernel, but this time the disk benchmarking is being done atop a high-performance solid-state drive courtesy of OCZ Technology and the CPU is an Intel Core i7. The drive being tested across these three leading file-systems is the OCZ Vertex 2 that promises maximum reads up to 285MB/s, maximum writes up to 275MB/s, and sustained writes up to 250MB/s.

    http://www.phoronix.com/vr.php?view=15187

  • #2
    Did you use the "ssd" mount option for btrfs?

    Comment


    • #3
      ZFS can use a flash drive as a cache device or intent log device.

      http://www.nerdblog.com/2010/03/zfs-...s-amazing.html
      http://blogs.sun.com/roch/entry/using_zfs_as_a_network

      For anyone considering a NAS or other multi-disk build with ZFS it is an interesting option.

      Comment


      • #4
        Originally posted by ernstp View Post
        Did you use the "ssd" mount option for btrfs?
        No, it's no longer needed or used. Btrfs will auto-detect if it's an SSD and apply optimizations accordingly. You can check in the dmesg when mounting Btrfs on an SSD and you should automatically see a message about SSD optimizations.
        Michael Larabel
        http://www.michaellarabel.com/

        Comment


        • #5
          Why not compare with the native ZFS performance in (open)Solaris instead of *BSD?

          Comment


          • #6
            Originally posted by wpoely86 View Post
            Why not compare with the native ZFS performance in (open)Solaris instead of *BSD?
            OpenSolaris b134 wouldn't boot on the ThinkPad W510.
            Michael Larabel
            http://www.michaellarabel.com/

            Comment


            • #7
              Does your kernels (or udev) detect your SSD disks as no rotational?

              (cat /sys/class/block/sda/queue/rotational)

              Comment


              • #8
                Looks like BSD+ZFS got murdered to me.

                Comment


                • #9
                  ext4 SSD tweaks?

                  Has anyone with an SSD tried using the RAID optimisation options in ext4 (also available in ext2/3) to tweak it for better SSD performance?

                  What I have in mind is that the underlying flash memory block size on an SSD could be considered to be analogous to the RAID stripe width in that modifying a smaller or non-aligned block of data results in a read-modify-write cycle. The '-E stripe-width=n' option to mke2fs tells the filesystem block allocator to place data so as to try to avoid read-modify-write cycles if possible (i.e. align it to the start of a block and fill an entire block wherever possible).

                  If it's possible to find out from an SSD manufacturer (or even by querying the drive?) the flash block size, it might be interesting to compare performance of a drive set up "any old how" with one containing a partition that is aligned to the start of a flash block bearing a filesystem created with the stripe width option. One would expect to see some difference in the write tests but not in the read tests.

                  Comment


                  • #10
                    Guys, thanks for your nice benchmarks on FSs, but you repeatedly ignore the CPU usage/system-load on such tests. There is a bottleneck here somewhere (Btrfs HDD vs SSD) which needs to be identified.

                    Comment


                    • #11
                      ...and another thing. Both ZFS and Btrfs have lots of features that are not present in EXT4. Some of them, such as indexing and dynamic volume management, can degrade performance. So knowing what is enabled and what is not, is also important. Does ZFS have more such features enabled by default etc?

                      Comment


                      • #12
                        It is very syntethic benchmark. In reality anyone using zfs will probably use many disks probably with configuration of at least two HDD and one SSD as secondary cache and intent log.

                        But benchmarking such configuration is also pretty hard. Benchmark would need many hours to run, to fill cache and see difference.

                        Anyway nice to see benchmark of modern file systems!

                        Comment


                        • #13
                          This is meant for desktop users. Just as mentioned in the rotating hdd benchmark thread, most users crying for btrfs or zfs and its features like snapshotting are actually single drive desktop and laptop. Yum's adoption of btrfs snapshotting is one good sign those kind of file systems will be used in ordinary desktops. There will probably be much more ordinary users using it on laptops than advanced users on multi-drive machines, just because there is so much more laptops out there.

                          Comment


                          • #14
                            Originally posted by misiu_mp View Post
                            This is meant for desktop users. Just as mentioned in the rotating hdd benchmark thread, most users crying for btrfs or zfs and its features like snapshotting are actually single drive desktop and laptop. Yum's adoption of btrfs snapshotting is one good sign those kind of file systems will be used in ordinary desktops. There will probably be much more ordinary users using it on laptops than advanced users on multi-drive machines, just because there is so much more laptops out there.
                            I hope. Yes. I use ZFS on laptop without any problem (using zfs-fuse). Snapshoting (from cron), simple dumping of snapshots differences to another ZFS system for backup, all content hashed, important metadata replicated 2-3 times (ditto blocks), documents and important projects also replicated for safety (2 times), everything compressed (gzip6, with exception of Music). It is really easy to setup and just works Hope btrfs will also have so good CLI (zpool and zfs commands just rocks).

                            On OpenSolaris there is excelent integration of ZFS features into the Nautilus on Gnome.

                            Comment


                            • #15
                              Congratulations to BTRFS. Personally, I dont care about speed. Is my data safe? That is the most important thing to me.

                              Would you prefer a cpu which is very fast, but does some miscalculations, or a slower cpu which always do correct calculations?

                              ZFS seems to be safe, according to researchers. XFS, JFS, ReiserFS, Raid5, Raid6, etc - all are not safe according to research. We see that it is very difficult to make safe storage solutions. There is no research on BTRFS data safety, so we have to wait for research. But now BTRFS is under development and under heavy revision where things are changed all the time, so we have to wait for finalization too.

                              Comment

                              Working...
                              X