[CP2K-user] CP2K performance in parallel run

[SS]

Dear CP2K users,

I was trying to run the AIMD simulation of my system containing 271 
atoms(metal complex and water molecules). I am not getting a scale-up in 
performance in parallel runs as I go from 8 to 16 and then to 32 
processors. It's taking almost the same time to complete the same number of 
MD steps.

I am using the following command

mpirun -np $np cp2k.popt  inp > out 

and I am running on an HPC cluster (CPU).

What could be the reason?
Expecting a reply

Thank you
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191104/bcfc96f2/attachment.htm>