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

  • #11
    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.
    Indeed you seem to point out the reason for the tests, and why we want them, Michael tests the default tool chains, on each distro, because that it usually what's in play when you choose a distro. In fact these very tests also highlight the differences you mentioned and the results attainable. These results show that the other distro's paid attention to the performance of Clear and started tweaking as required. Even ubuntu are following suite. All thanks to these type of benchmarks.

    Now as for your request (feel free to premium up if you haven't already) it seems that in terms of what you want it might be an idea to test snap or flatplak et al (the other one) across each and see the performance. Seems like the most reasonable option for automated benchmarks. So suggest some available snap /flat pak etc.

    Comment


    • #12
      Originally posted by kgardas View Post
      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.
      On OpenBSD you need to enable url fopen setting from its php.ini file as it ships with that support disabled for its PHP package.
      Michael Larabel
      http://www.michaellarabel.com/

      Comment


      • #13
        Originally posted by microcode View Post
        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.
        Not true! OpenBSD FFS syncs only meta-data write by default. Data write is async!

        Comment


        • #14
          Originally posted by Michael View Post

          On OpenBSD you need to enable url fopen setting from its php.ini file as it ships with that support disabled for its PHP package.
          Thanks! This indeed helps!

          Comment


          • #15
            Michael, thanks for testing OpenBSD! I'm glad it passed in all tests. The slowness in some of those is caused by synchronous meta-data write. I've been surprised that even "read" test of BlogBench still writes a lot of data. I'm on rotating rusty WD black 2.5" drive and with softdep mounted dir I've been able to beat your read on NVMe more than twice. So yes, intensive meta-data write on OpenBSD sucks. Good thing is that this is solvable by softdep (or hopefully in some future by WAPBL).

            Comment


            • #16
              My experience with Clang/LLVM is that it produces significantly subpar code for x86-64, compared to gcc (clang/llvm was primarily meant for ARM, where it does a better job than gcc, IMO).

              I'm willing to bet that some of the BSD benchmarks would benefit considerably from being built with gcc.

              Comment


              • #17
                Originally posted by vladpetric View Post
                I'm willing to bet that some of the BSD benchmarks would benefit considerably from being built with gcc.
                NetBSD and DragonFly have GCC for system compiler. Theory should be easy enough to check.,

                Comment


                • #18
                  Just be aware that OpenBSD 6.4, disables hyperthreading by default, so the CPU comparisons might be very well flawed.

                  Comment

                  Working...
                  X