I have a machine which i can boot into either Ubuntu 18.04.5 or Fedora 32. I have had some performance issues, so i thought i would run a quick benchmark on both distros. I installed sysbench from the package manager and ran the CPU and memory benchmarks, for which i noted the events per second and total operations per second respectively. The machine has four physical cores, so i ran with 1 and 4 threads.

Ubuntu has sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3).

Fedora has sysbench 1.0.17 (using system LuaJIT 2.1.0-beta3).

The version and the results are the only difference in the output on the two distros; the options like number of primes, block size, etc, are the same.

The results are:

Distro          CPU (1 thr)  CPU (4 thr)  memory (1 thr)  memory (4 thr)
Ubuntu 18.04.5      1213.05      4384.59      5332997.51      8069089.73
Fedora 32         156442.30    564670.95      6220026.17      9055910.28

The differences in the memory results seem plausible, but the differences in the CPU results are rather surprising.

Should i expect these to be more similar?


Points of interest:

  • Packages versions are different enough, the new version might contain optimizations not found in the old one

  • Most Linux distros compile packages differently, i.e. not only using different compilers but different compiler optimizations/flags - they might affect your scores as well.

  • Most likely you can run Ubuntu's sysbench in Fedora directly without any issues - please do and check the results. If they mirror your Ubuntu results that will confirm either or both of the previous points.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.