Announcement

Collapse
No announcement yet.

Benchmarks Of The New ZFS On Linux: EXT4 Wins

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

  • Benchmarks Of The New ZFS On Linux: EXT4 Wins

    Phoronix: Benchmarks Of The New ZFS On Linux: EXT4 Wins

    At the end of March was a new release of ZFS On Linux, a kernel module implementation of the ZFS file-system for Linux, and it was declared ZFS On Linux is now ready for wide-scale deployments. With this release (ZOL/SPL v0.6.1), new benchmarks are being done to compare ZFS to popular Linux file-systems. In this article is a brief preview against EXT4...

    http://www.phoronix.com/vr.php?view=MTM1NTA

  • #2
    Originally posted by phoronix View Post
    Phoronix: Benchmarks Of The New ZFS On Linux: EXT4 Wins

    At the end of March was a new release of ZFS On Linux, a kernel module implementation of the ZFS file-system for Linux, and it was declared ZFS On Linux is now ready for wide-scale deployments. With this release (ZOL/SPL v0.6.1), new benchmarks are being done to compare ZFS to popular Linux file-systems. In this article is a brief preview against EXT4...

    http://www.phoronix.com/vr.php?view=MTM1NTA
    What about a comparison against OpenSolaris derivatives? In my opinion, this benchmark is totally flawed without it.

    Comment


    • #3
      Well, ryao won't be pleased with this article in the slightest.

      Comment


      • #4
        Isn't this like comparing an armoured truck versus a sportscar with central locking?

        Comment


        • #5
          From the ZFS on Linux FAQ:
          Additionally, it should be made clear that the ZFS on Linux implementation has not yet been optimized for performance. As the project matures we can expect performance to improve.

          Comment


          • #6
            Originally posted by xeekei View Post
            Isn't this like comparing an armoured truck versus a sportscar with central locking?
            come on ! at least enable checksumming for everything (where possible) and some other fancy stuff for ext4

            then it's at least partly comparable



            raw performance-wise:

            when backing up my 1.2-1.5 TB of data from an ext4 home-partition to ext4 vs. btrfs & ZFS

            btrfs (with gzip-compression) takes about 2x the time than ext4 and approx. 1.5x the time of XFS (strangely xfs got slower with my latency-related tuning & ext4 got faster )

            ZFS with lz4 and/or lzjb compression takes about 2.5x-3x the time of ext4


            and that's only the raw transfer-time (8 minutes of ext4, approx. 30 minutes of ZFS) and neglecting all the data-safety wise features of e.g. ZFS:


            - sha256 checksums (said to be the slowest compared the default hash algorithm, lower probability of hash collisions with e.g. deduplication)
            - checksums on the whole filesystem - from file, metadata up to higher parts of the filesystem
            - optional single-filesystem dedup & multi-copy functionality (didn't know about that until recently and will use it on critical study/personal files): e.g. copies=2
            - there's several more - and I'm not really an expert in filesystems so just let your google-fu take you to wikipedia, etc. to find out more

            Originally posted by Vim_User View Post
            From the ZFS on Linux FAQ:

            Additionally, it should be made clear that the ZFS on Linux implementation has not yet been optimized for performance. As the project matures we can expect performance to improve.

            looking forward to it

            Comment


            • #7
              Originally posted by GreatEmerald View Post
              Well, ryao won't be pleased with this article in the slightest.
              These tests are invalid. I suspect that they were cherry-picked, although I cannot prove that.

              The tests were done with early Intel SSDs that are known to lie about their block size, presumeably without making any adjustment to ashift. ext4's assumption that everything is 4KB is advantageous here. In additional, 3 of the 4 tests appear to be read/write once tests, which do not reflect real world workloads. I say 3 of 4 because I have not read enough about dbench to say one way or another, although I can say that the single client configuration is likely the worst possible test possible.
              Last edited by ryao; 04-18-2013, 06:05 PM.

              Comment


              • #8
                Originally posted by kernelOfTruth View Post
                come on ! at least enable checksumming for everything (where possible) and some other fancy stuff for ext4

                then it's at least partly comparable



                raw performance-wise:

                when backing up my 1.2-1.5 TB of data from an ext4 home-partition to ext4 vs. btrfs & ZFS

                btrfs (with gzip-compression) takes about 2x the time than ext4 and approx. 1.5x the time of XFS (strangely xfs got slower with my latency-related tuning & ext4 got faster )

                ZFS with lz4 and/or lzjb compression takes about 2.5x-3x the time of ext4


                and that's only the raw transfer-time (8 minutes of ext4, approx. 30 minutes of ZFS) and neglecting all the data-safety wise features of e.g. ZFS:


                - sha256 checksums (said to be the slowest compared the default hash algorithm, lower probability of hash collisions with e.g. deduplication)
                - checksums on the whole filesystem - from file, metadata up to higher parts of the filesystem
                - optional single-filesystem dedup & multi-copy functionality (didn't know about that until recently and will use it on critical study/personal files): e.g. copies=2
                - there's several more - and I'm not really an expert in filesystems so just let your google-fu take you to wikipedia, etc. to find out more




                looking forward to it
                You should use zfs send/recv for backups. It will outperform rsync on ext4, especially when doing incremental backup.

                Comment


                • #9
                  Feature wise comparison

                  I find it a little bit useless to just compare ZFS against EXT4.
                  EXT4 is interesting as a point of comparison. But they don't even belong to the same generation.

                  EXT4 is basically EXT3 with extents added in (which in turn was EXT2 with journalling thrown in).
                  It an previous-gen filesystem.

                  Where as ZFS is a newer generation file system with a ton of features (with checksumming of nearly everything, data de-duplication, redundancy, on-the-fly compression, copy-on-write, subvolumes and snapshots, and a whole volume management).
                  It should be compared against similarly full-of-feature files system (btrfs more or less covers the same feature set).

                  ZFS, BTRFS and Co vs. EXT4 and Co are never going to run at the same speed, given that the first have much more features that require some CPu cycles. And they don't target the same audience.


                  The current benchmark would be like benchmarking FAT32 vs. a distributed cluster file system. Even if they are all filesystems, they don't have the same feature set at all and don't aim for the same users.

                  Comment


                  • #10
                    Originally posted by crapI3SD
                    and you are sane
                    Freudian slip?
                    Seriously, why hasn't this guy (ryao) been kick from the gentoo project.
                    Maybe because, contrary to you, he is able to contribute to the open source world?

                    Comment


                    • #11
                      wow, what posts... one bullshit post after the other. even kindergarden would have a higher educated level than what i read here.

                      Comment


                      • #12
                        asdx, as always inconsistent with his own opinion:
                        Originally posted by asdx
                        I really don't care if BSD or Linux shares ideas between them, in fact, I believe that's a good thing.
                        So actually you think that ZFS on Linux is a good thing? So why do you now state it the other way around? By the way, the same thing that is done by the antiBSD troll, in his previous incarnations as "systemd anals BSD", as you can see here (the original post was deleted): http://phoronix.com/forums/showthrea...280#post309280

                        Grow up, start to contribute instead of bitching around on forums and maybe start to be consistent with your own opinion, then people will start to take you seriously.

                        Comment


                        • #13
                          Jesus, don't feed the trolls, people. Seriously, the Phoronix forums are disgraceful. Spamming the forums with crap like this really should get you banned.

                          I will say though that I would expect ZFS to be slower that EXT4, but there are no details here on how either filesystem was configured, and unless ashift for ZFS was set correctly these results are invalid. With that said, you can expect performance for SSD-backed ZFS to suck pretty hard after a full disk of writes, since there's still no vdev TRIM support. For closer apples to apples you could also configure both filesystems to have more comparable features enabled, but whatever.

                          Multi-disk comparisons should be interesting, though I don't know how you decide which pool topologies you test, and what you do about ZFS support for SSD-backed ZIL and L2ARC.
                          Last edited by pdffs; 04-18-2013, 08:59 PM.

                          Comment


                          • #14
                            Originally posted by pdffs View Post
                            Multi-disk comparisons should be interesting, though I don't know how you decide which pool topologies you test, and what you do about ZFS support for SSD-backed ZIL and L2ARC.
                            An interesting test, and the only one that really matters in the real world, would be which one loses data quicker. Ext4 seems to have the lead there.

                            Comment


                            • #15
                              I love reading comments from grown men acting like prebubescent punks on the tech sites. It reminds me of 3rd grade when the fat kid beat up on the smallest kid in school. Later on as they both matured the fat kid became the runt and ostracised from progress. The same will be with filesystems.

                              Comment

                              Working...
                              X