Announcement

Collapse
No announcement yet.

Windows Server 2016 vs. FreeBSD 11.2 vs. 8 Linux Distributions Performance

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

  • Windows Server 2016 vs. FreeBSD 11.2 vs. 8 Linux Distributions Performance

    Phoronix: Windows Server 2016 vs. FreeBSD 11.2 vs. 8 Linux Distributions Performance

    Given the recent releases of FreeBSD 11.2, Scientific Linux 6.10, openSUSE Leap 15, and other distribution updates in the past quarter, here are some fresh benchmarks of eight different Linux distributions compared to FreeBSD 11.2 and Microsoft Windows Server 2016. The tested Linux platforms for this go-around were CentOS 7.5, Clear Linux 23610, Debian 9.4, Fedora Server 28, openSUSE leap 15.0, Scientific Linux 6.10, Scientific Linux 7.5, and Ubuntu 18.04 LTS.

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    I'm thrilled Windows consistently takes a beating, but surprised too. I wonder what Microsoft is doing wrong technically speaking (without regards to their evil actions and business model) to perform so poorly.

    In the free time you don't have, Michael, it might be interesting to throw some kind of C# benchmark into the mix. Maybe Windows will take a lead there as the native platform for the code.

    Comment


    • #3
      ATTENTION PLEASE, general service announcement:

      FreeBSD's performance in these tests is very likely caused by ZFS (the default filesystem), which is a CoW filesystem and as such it does have worse performance than a journaled filesystem (unless you go out of your way to add SSD caching and such).

      Let's not start flaming FreeBSD plz, keep it classy, flame Windows server only.

      Comment


      • #4
        Originally posted by Michael_S View Post
        In the free time you don't have, Michael, it might be interesting to throw some kind of C# benchmark into the mix. Maybe Windows will take a lead there as the native platform for the code.
        Patches of new test profiles are also happily accepted Or anyone that is a C# domain expert that can put together at least a nice test script of some suitable C# benchmarks. I've only ever had to dabble briefly with C# and not familiar with what good C# benchmarks may or may not be out there and suitable for testing.
        Michael Larabel
        https://www.michaellarabel.com/

        Comment


        • #5
          Originally posted by Michael_S View Post
          I'm thrilled Windows consistently takes a beating, but surprised too. I wonder what Microsoft is doing wrong technically speaking (without regards to their evil actions and business model) to perform so poorly.
          Taking a guess... they aren't really investing as much as they should on that, as they don't need to.
          They just rely on vendor lockdown, or inertia from legacy software/developer companies, or similar things where they don't need to actually show they run faster than Linux.

          Comment


          • #6
            Originally posted by starshipeleven View Post
            ATTENTION PLEASE, general service announcement:

            FreeBSD's performance in these tests is very likely caused by ZFS (the default filesystem), which is a CoW filesystem and as such it does have worse performance than a journaled filesystem (unless you go out of your way to add SSD caching and such).

            Let's not start flaming FreeBSD plz, keep it classy, flame Windows server only.
            AFAIK, opensuse uses BTRFS by default. But it could be that in some cases CoW is disabled, IIRC on /var, to enhance database and vm performance.

            Does anybody know about ReFS on windows? I guess it's not used by default ATM?

            Comment


            • #7
              Originally posted by Michael_S View Post
              I'm thrilled Windows consistently takes a beating, but surprised too. I wonder what Microsoft is doing wrong technically speaking (without regards to their evil actions and business model) to perform so poorly.

              In the free time you don't have, Michael, it might be interesting to throw some kind of C# benchmark into the mix. Maybe Windows will take a lead there as the native platform for the code.
              They don't care and they don't need. What I see in my country, they have the monopoly, the teachers are old and only use windows, etc.

              Comment


              • #8
                Originally posted by andre30correia View Post
                They don't care and they don't need. What I see in my country, they have the monopoly, the teachers are old and only use windows, etc.
                On servers? Really?

                I see Windows everywhere, literally... on desktops.

                ...never once seen a server with Windows on it.

                Not doubting you but yeah.

                Comment


                • #9
                  Originally posted by treba View Post
                  AFAIK, opensuse uses BTRFS by default. But it could be that in some cases CoW is disabled, IIRC on /var, to enhance database and vm performance.
                  Yes on OpenSUSE they disable CoW in some paths, which is why it is not performing bad on every metric.

                  Does anybody know about ReFS on windows? I guess it's not used by default ATM?
                  It's not. They are taking some steps to increase its popularity, but the fact that you can't boot from a ReFS volume is a major show-stopper (and quite frankly annoying as it is a 100% artificial limitation, it's not like it's hard for them to add this support).

                  Comment


                  • #10
                    Originally posted by starshipeleven View Post
                    Taking a guess... they aren't really investing as much as they should on that, as they don't need to.
                    They just rely on vendor lockdown, or inertia from legacy software/developer companies, or similar things where they don't need to actually show they run faster than Linux.
                    I understand that, but considering the extra money they get selling Windows Server licenses, and the higher cost of using Windows on all of the cloud providers including Windows Azure, I think they would want to compete better.

                    Comment

                    Working...
                    X