Announcement

Collapse
No announcement yet.

The Performance Of Clear Linux vs. Fedora vs. Ubuntu Over 2020

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

  • The Performance Of Clear Linux vs. Fedora vs. Ubuntu Over 2020

    Phoronix: The Performance Of Clear Linux vs. Fedora vs. Ubuntu Over 2020

    Earlier this week we looked at the performance of Intel's Clear Linux over the past year but how does that compare to the likes of say Fedora and Ubuntu? This article is looking at the performance of Fedora Workstation, Ubuntu, and Clear Linux on the same hardware over the past year.

    http://www.phoronix.com/vr.php?view=29861

  • #2
    Sad Fedora is still belonging to the underperforming distributions, even with LTO enabled with Fedora 33.

    Comment


    • #3
      Just a friendly reminder: if you are on Arch or any of its derivatives, it's trivial to install a patched kernel from TkG, namely "linux-tkg", which includes patches from Clear Linux.

      Comment


      • #4
        linux-tkg technically supports non-Arch distributions, but they're likely second-class.

        Comment


        • #5
          I wonder what specifically is allowing clear linux to have such large gains in things like dav1d, it getting basically double the performance.

          Comment


          • #6
            The Go benchmarks are particularly interesting as they are unaffected by the CC,CXX compiler flags, thus any improvement in performance there should mean the underlying system is faster, rather than just the benchmarked programs running faster due to being compiled with higher optimization.

            Any chance you'll add more Go benchmarks in future comparisons ?

            Comment


            • #7
              Could have just linked the README with instructions for most RPM and DEB based distributions.

              Comment


              • #8
                Originally posted by Grinch View Post
                The Go benchmarks are particularly interesting as they are unaffected by the CC,CXX compiler flags, thus any improvement in performance there should mean the underlying system is faster, rather than just the benchmarked programs running faster due to being compiled with higher optimization.

                Any chance you'll add more Go benchmarks in future comparisons ?
                There's a few Go benchmarks in PTS but not aware of too many real-world applicable Go benchmarks / popular Go workloads with nice representative benchmark cases?
                Michael Larabel
                http://www.michaellarabel.com/

                Comment


                • #9
                  I'll have to try it out. My inspiron has always failed the compatibility check in the past, but today it passes. It needs a fast os.

                  Comment


                  • #10
                    Originally posted by 900k View Post
                    I wonder what specifically is allowing clear linux to have such large gains in things like dav1d, it getting basically double the performance.
                    At the same time I wonder how relevant dav1d 0.5 is at the end of 2020, when dav1d is already at 0.8.

                    Originally posted by Grinch View Post
                    The Go benchmarks are particularly interesting as they are unaffected by the CC,CXX compiler flags, thus any improvement in performance there should mean the underlying system is faster, rather than just the benchmarked programs running faster due to being compiled with higher optimization.

                    Any chance you'll add more Go benchmarks in future comparisons ?
                    I don't think Go compilation uses gcc, so those flags won't work.

                    Originally posted by Michael View Post
                    There's a few Go benchmarks in PTS but not aware of too many real-world applicable Go benchmarks / popular Go workloads with nice representative benchmark cases?
                    The docker ecosystem is heavily dependent on Go, but I imagine building tests around Docker or Kubernetes isn't that straightforward.
                    Besides that, maybe this is something that can be benchmarked more easily: https://github.com/future-architect/vuls

                    Comment

                    Working...
                    X