Announcement

Collapse
No announcement yet.

Bare Metal Benchmarking Alpine Linux 3.11 Against Ubuntu 19.10 + Clear Linux

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

  • Bare Metal Benchmarking Alpine Linux 3.11 Against Ubuntu 19.10 + Clear Linux

    Phoronix: Bare Metal Benchmarking Alpine Linux 3.11 Against Ubuntu 19.10 + Clear Linux

    While Alpine Linux has traditionally been a lightweight Linux distribution focused on use within containerized environments, with yesterday's release of Alpine Linux 3.11 brought GNOME and KDE support for those wanting to use this distribution as a desktop/workstation OS. Curious, I had to give it a try and of course run some general Alpine Linux 3.11 benchmarks up against Clear Linux and Ubuntu 19.10 for seeing how its performance is on bare-metal hardware.

    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
    Alpine is actually very end user friendly...its just more picky about its users

    Personally I hope they don't loose focus and try to be a desktop distro with a 1 click installation.
    There are enough of them already, all doing the same thing...

    I would like to see them become "the mother distro" for some spin-offs like debian is today.
    Postmarket-OS is already one.
    Maybe Postmarket-OS will expand their horizon to include desktops/laptops, they already have most of the infrastructure in place to make a desktop distro.
    Last edited by Raka555; 20 December 2019, 10:54 AM.

    Comment


    • #3
      Surprisingly good. Isn't musl mostly developed by a single author? Some also claim that Alpine produces smaller installation footprint than Ubuntu.

      Comment


      • #4
        Originally posted by caligula View Post
        Surprisingly good. Isn't musl mostly developed by a single author? Some also claim that Alpine produces smaller installation footprint than Ubuntu.
        There's no "claim" about it. When Alpine docker images started to be produced, they were around one third the size of an Ubuntu base image. Full desktop installs are also smaller, though less so as you factor in the packages drawn in as dependencies. Musl is designed to produce more compact binaries than glibc. AddionalyA the smaller binary requires less cache in the cpu, giving a minor speedup, as you see in the charts.

        Comment


        • #5
          I wonder why alpines performance isn't better compared to Ubuntu?
          I thought musl is much faster than glibc?
          And systemd trolls tried to tell me that systemd is a bloated and slow piece of software too.

          Anyone knows the reason?

          Comment


          • #6
            Originally posted by flower View Post
            I wonder why alpines performance isn't better compared to Ubuntu?
            I thought musl is much faster than glibc?
            And systemd trolls tried to tell me that systemd is a bloated and slow piece of software too.

            Anyone knows the reason?
            The "claim" is musl is more standards compliant and less bloated, not necessarily faster.
            Also these benchmarks were done on a high end CPU with a lot of memory.
            The picture might look different on something like an RPI with 512M/1G RAM. (hint, hint Michael)

            None of the benchmarks here had anything to do with the init system, so you wont see a difference based on that.

            See Snaipersky's post above.
            Last edited by Raka555; 20 December 2019, 12:59 PM.

            Comment


            • #7
              Alpine also don't get their panties in a knot about supporting 32bit...

              Comment


              • #8
                Originally posted by flower View Post
                I wonder why alpines performance isn't better compared to Ubuntu?
                I thought musl is much faster than glibc?
                And systemd trolls tried to tell me that systemd is a bloated and slow piece of software too.

                Anyone knows the reason?
                Ubuntu/Debian have had plenty of time and resources to fix performance issues, glibc probably had some speed tricks helping out, whereas musl is much more bare bones. Furthermore, larger caches somewhat mitigate the code size difference.
                Michael, could we trouble you for a second round of tests on ryzen 3rd gen?

                Comment


                • #9
                  Originally posted by flower View Post
                  I wonder why alpines performance isn't better compared to Ubuntu?
                  I thought musl is much faster than glibc?
                  And systemd trolls tried to tell me that systemd is a bloated and slow piece of software too.

                  Anyone knows the reason?
                  Stop listening to the trolls and fanbois? systemd does not affect performance of applications or system daemons in any way (negative or positive) and glibc contains tons of hand optimized algorithms for tons of architectures.

                  Comment


                  • #10
                    Originally posted by caligula View Post
                    Surprisingly good. Isn't musl mostly developed by a single author? Some also claim that Alpine produces smaller installation footprint than Ubuntu.
                    An Alpine Linux LXC container is around 10mb in size - an Ubuntu container is 300-400mb.

                    This is why Alpine is the default flavor of Docker.

                    Comment

                    Working...
                    X