0

I`m running a database benchmark (Tpc-h), and running a same SQL query several times. To my surprise, the time spent has a significant difference, even though it is the same query, reaching up to 40% or 50% of the difference between one execution and another.

I'm using a server HP MODEL: Proliant ML110-G9 Processador: (1) Intel Xeon E5-1603v3 (2.8GHz/4-core/10MB/140W). In this way I did according to the UEFI(BIOS) manual (page 65), setting "Power Management" to "Maximum Performance". https://support.hpe.com/hpsc/doc/public/display?docId=c04398276

I re-executed the benchmark, but it did not work, because the time spent has a significant difference yet. My question is if you have any additional settings that I have to do in the BIOS yet? Another question is if I have to configure the "Performance" mode also in the Operating System. I use Debian 8 (Jessie).

Best Regards

Neto

  • If you configure CPU power management in the BIOS, then OS control will not be possible. Perhaps your performance issue is with something other than CPU. – Michael Hampton Nov 10 '17 at 1:56
  • 1
    The CPU is only one component of performance. Without any insight in to your query, your database, and other metrics and tests, it's impossible to suggest why you experience this behavior. Any number of other activities or bottlenecks could be coming in to play. – Appleoddity Nov 10 '17 at 3:58
  • I agree with @Appleoddity and Michael. Have a look at other possible bottlenecs on your system: superuser.com/questions/1264798/why-wont-my-cpu-operate-at-its-max-potential-even-when-my-application-which-ut/1264808#1264808 Also give us some overview of your measuring method you use. – Jaroslav Kucera Nov 10 '17 at 8:14

Your Answer

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

Browse other questions tagged or ask your own question.