I see that after reboot, my first parallel simulation has 12% better performance than all the other ones after. This seems to come from the CPU cache.
On Linux, the documentation from AMD advise to fluch the cache using the following commands
sync; echo 3 > /proc/sys/vm/drop_caches --> to drop all caches
-e MPI_FLUSH_FCACHE=1, full,v --> to drop mpi caches
However, I do not find any equivalent on Windows 10 Pro. My application does not run on Linux, so I am wondering if there is any way to get the performance of the first simulation on all (without rebooting between each simulation...)?
Thanks in advance for your suggestions!