Energy Measurements

Energy Measurements

by Sudesh Rathnayake -
Number of replies: 0

Hello all,

I am using likwid-perfctr to measure energy of my OpenMP code (the 2D version in git@gitlab.cs.fau.de:yh54ojyn/mucosim.git). However, as shown in the attached figure, there is a strange energy measurement result at 1st core.


I'm using the likwid-perfctr command as srun --cpu-freq=2400000-2400000:performance likwid-perfctr -C S0:0 -g ENERGY-m ./CLBM


I have taken the measurements several times, and it is the same as this. However, when I get the DRAM energy results, there is no deviation and I cannot observe the

strange behavior.


Could you please give any comment regarding this variation?