New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
The same configuration but different CPU scores?
I own two servers with identical configurations and virtualization setups, but when testing the CPU score of VPS (with the same operating system) running on the two servers, they yield different results. I have checked the settings and they are all the same, except for the version of QEMU.
Additionally, I have a server and want to virtualize KVM, what would be the best platform to run KVM VPS? Thank you very much and looking forward to your suggestions.
Comments
The load on a server isn't static, so it's normal that there is a small difference in the test result each time you run a test.
If you test with yabs.sh / Geekbench. What are the scores?
Are the dedicated servers idling?
How much different?
Your bare metal servers? Or VPS servers that you pay for?
about a half, quite different between my servers
Yes my bare server
it is not the 'identical' setup then.
check bios config maybe, hyperthreading/boost enabled or cooling/throttling or other settings that might have the CPU behave differently.
in each layer, there could be som things hidden. however your request lacks in detail, so all everyone can do is just guessing around.
Do you observe the same CPU frequencies under full utilization for both servers? Your servers might have different cooling setups or your benchmark is not reproducible enough.
Curious about something.. If they are identical servers, why are they running different versions?
Otherwise, while they may have identical hardware, they don't have identical settings (or a fan or two has failed).
Normal troubleshooting applies...
Check that they actually have the same CPU and nothing else running while you test.
After that, check all the BIOS options, are they all the same, especially cooling and performance settings.
Monitor temperatures and CPU utilization while testing.
There has to be only 1 running VPS on a bare-metal server during the test. For virtualization, I would suggest Proxmox.
Check:
Power Settings to be the same, especially on CPU.
OS version should be the same on both servers.
Memory MHz and timing should be the same.
BIOS + ILO/iDRAC/BMC should be the same.
If all the above mach, you can still get a +/- 5% difference in performance.
PS:
for the moment as @vsys_host said, we also recommend PROXMOX.