Announcement

Collapse
No announcement yet.

Initial Benchmarks Of OpenBSD 6.4, DragonFlyBSD 5.3, FreeBSD vs. Linux

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

  • Initial Benchmarks Of OpenBSD 6.4, DragonFlyBSD 5.3, FreeBSD vs. Linux

    Phoronix: Initial Benchmarks Of OpenBSD 6.4, DragonFlyBSD 5.3, FreeBSD vs. Linux

    Given the recent release of OpenBSD 6.4, FreeBSD 12 now being in beta, and DragonFlyBSD 5.3 evolving nicely for what will eventually ship as DragonFlyBSD 5.4, here is the start of some fresh benchmarks between the BSDs and a few Linux distributions to see how the performance compares as we approach the end of 2018.

    http://www.phoronix.com/vr.php?view=27046

  • #2
    In before flamewar..

    Comment


    • #3
      I must say I find these kinds of benchmarks rather pointless. There are simply too many variables, it's posed as a operating system comparison, but what is actually benchmarked is a program compiled with one optimization level against a program with another optimization level, it says nothing about the performance of the underlying system.

      Please do a benchmark where you use the same toolchain or at the very least the same optimization level (preferably the highest) on the tests, as any differences in results would then be represented by the performance of the actual operating systems the tests run on.

      In my opinion there's practically nothing you can learn from a test such as this one presented here, on one system the benchmark was compiled with -O2, on another -O3 -march native, on one system it used clang/llvm, on another GCC, etc. It's a compiler options/toolchain comparison masquerading as a operating system/distro comparison. It really doesn't make much sense to me.

      Comment


      • #4
        Typo:

        Originally posted by phoronix View Post
        It will be interesting to to run this test

        Comment


        • #5
          Michael, any trick to run phoronix test suite on OpenBSD? I can reach all sites with ping, but `phoronix-test-suite network-info` tells me on all "Can Reach" simple "NO".
          BTW, I'm not behind the proxy and also phoronix seems to run fine inside CentOS7 which is running in VMM on top of the tested OpenBSD. So the problem is probably in OpenBSD's php and its inability to reach your hosts. How have you solved that? I'm on 6.4-current though.
          Last edited by kgardas; 11-02-2018, 12:33 PM. Reason: note about proxy and centos

          Comment


          • #6
            Wow the results are all over the place, seems like clear linux still generally reigns supreme though, but I have to admit the performance of the BSDs is quite impressive.

            Comment


            • #7
              Hey Michael, I just had this idea that maybe you should try to get the best average performer in these benchmarks. It shouldn't be that hard, considering you can just tally up the total score for each test where more is better, and then substract from that total everything where less is better. Maybe create something like a 'final score' for each platform/hardware tested, at the end of the benchmark.

              Sometimes the platform that wins most often might not be the best average performer; maybe some other platform is always in second-third place or something giving it better results on average even if it rarely if ever won. While the number of wins is quite interesting, I think this too would be a useful metric.

              Comment


              • #8
                Originally posted by Grinch View Post
                I must say I find these kinds of benchmarks rather pointless. There are simply too many variables, it's posed as a operating system comparison, but what is actually benchmarked is a program compiled with one optimization level against a program with another optimization level, it says nothing about the performance of the underlying system.
                I disagree, I think having benchmarks like that is very useful.

                Sure, digging in in detail which exact component makes what difference gives you the real info (and regular Phoronix articles do that already), but it's quite useful to sample at the highest level so that you can see what's possible and what the various components _are_ that you should investigate in detail.

                Comment


                • #9
                  Originally posted by Grinch View Post
                  I must say I find these kinds of benchmarks rather pointless. There are simply too many variables, it's posed as a operating system comparison, but what is actually benchmarked is a program compiled with one optimization level against a program with another optimization level, it says nothing about the performance of the underlying system.

                  Please do a benchmark where you use the same toolchain or at the very least the same optimization level (preferably the highest) on the tests, as any differences in results would then be represented by the performance of the actual operating systems the tests run on.

                  In my opinion there's practically nothing you can learn from a test such as this one presented here, on one system the benchmark was compiled with -O2, on another -O3 -march native, on one system it used clang/llvm, on another GCC, etc. It's a compiler options/toolchain comparison masquerading as a operating system/distro comparison. It really doesn't make much sense to me.
                  While you are quite correct, there is also this fact that people will mostly use the supplied software so people who choose to run OpenBSD or Fedora will run applications/servers compiled with whatever flags, compilers and tool chains that the rest of the system uses. But that is not to say that it would be interesting to see say apache/postgresql compiled exactly the same on the various systems to see how well the various kernels and IO systems scale.

                  Comment


                  • #10
                    I guess OpenBSD + SQLite works well since OpenBSD always syncs writes by default, which is what SQLite takes extra effort to do on the other platforms.

                    Comment

                    Working...
                    X