Announcement

Collapse
No announcement yet.

Linux vs. BSD CPU Scaling Up To 20 Threads On The Core i9 7900X

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

  • #11
    Originally posted by darkfires View Post
    I can't believe you labeled this article Linux vs. BSD when most of the tests don't even have BSD in them and the ones that do have various different versions. Pretty much pointless.
    I know for a fact john the ripper easily compiles on BSD, and with some minimal effort you could easily get 99% of the tests working.
    You also need to configure powerd on BSD for turbo to work.
    I also found it disappointing that most of the tests didn't have all of the BSD operating systems working plus there were several main BSD versions missing (net and open). It would also be nice to see at least one OS that uses the Illuminos kernel for a more complete comparison. Perhaps as a suggestion, since the BSDs with the exception of perhaps TrueOS require more hands on configuration to set up and fine tuning are are not just set it up and benchmark like Fedora or Ubuntu a spare SSD could be set up for a given system with each OS on it ready to go for benchmarking. Perhaps even two sets of SSDs for each OS, one very generic that should boot on any cpu and one optimized to kingdom come for a specific cpu with a CFLAG optimizations like Clear Linux for a more accurate apples to apples comparison, plus it would save OS install time for each test.

    Just a suggestion though, great work as always for a one man team, appreciate any Linux vs BSD vs Solaris vs Other comparison because there aren't many other options out there!

    Comment


    • #12
      Very interesting benchmark. Thank you!

      Comment


      • #13
        The OpenMPI CFD solver results show starting to hit the point that all CFD solvers eventually do of 'parallel slowdown', where the amount of inter-process communication workload has increased to the point it out-weights any gains. The gains for CFD solvers from hyperthreading are generally known to be very marginal, so 10c/20t is more likely to show that effect than 20c.

        Comment


        • #14
          It will be great to see glibc benchmarks with a per-thread cache enabled.

          Comment


          • #15
            Originally posted by Pawlerson View Post
            It will be great to see glibc benchmarks with a per-thread cache enabled.
            Clear Linux currently has that patch backported... and the version in these benchmarks has that.

            Comment


            • #16
              Originally posted by darkfires View Post
              I can't believe you labeled this article Linux vs. BSD when most of the tests don't even have BSD in them
              because bsd is a joke and lost before start

              Comment


              • #17
                Originally posted by pal666 View Post
                because bsd is a joke and lost before start
                So how about not doing this kind of snark?
                I have great respect for anyone else who puts an OS together, be it a BSD variant or a Linux distro. Doing this is a lot of work, often thankless and all you get is bugs and complaints...
                Anyone doing that deserves at least that much respect. You can think one is better than the other, and each OS will have their strengths and weaknesses, but hard work it is.

                Comment


                • #18
                  Originally posted by Steffo View Post
                  I would also be interested on a comparison with Win 10
                  Win 10 might have a bottleneck in process destruction ATM (such as during compilation) that also affects the user interface:
                  This story begins, as they so often do, when I noticed that my machine was behaving poorly. My Windows 10 work machine has 24 cores (48 hyper-threads) and they were 50% idle. It has 64 GB of RAM an…

                  Comment


                  • #19
                    well same here, why does freeBSD 11 do not appear in half of the tests ?

                    Comment


                    • #20
                      Originally posted by Oppenheimer View Post
                      The OpenMPI CFD solver results show starting to hit the point that all CFD solvers eventually do of 'parallel slowdown', where the amount of inter-process communication workload has increased to the point it out-weights any gains. The gains for CFD solvers from hyperthreading are generally known to be very marginal, so 10c/20t is more likely to show that effect than 20c.
                      It is not uncommon for hyperthreading to be turned off on a scientific workstation. As noted sometime the gains just aren't there, sometimes you even see regressions.

                      Comment

                      Working...
                      X