[CP2K-user] Speeding up AIMD simulation

ASSIDUO Network lenardc... at gmail.com
Wed Mar 31 04:55:49 UTC 2021


Hi there everyone, I hope you are all doing well.

I'm trying to do an AIMD simulation of 6 layers of Au (111) (3 fixed) with 
CO and O adsorbed on top of the gold surface. I have the optimized geometry 
for this system (gold.xyz). Can I have some input on speeding up the 
equilibration, since simulating it with 32 CPUs and 4 GPUs over 12 hours 
does not even get me past 20 fs.

I know removing the &KPOINTS section would speed up the simulation, but I'm 
going after accuracy here. Do you have any suggestions on how to speed up 
my simulation? I have included my input file.

Kind regards,
Lenard
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210330/cb9b363f/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gold.inp
Type: chemical/x-gamess-input
Size: 4108 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210330/cb9b363f/attachment.inp>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gold.xyz
Type: chemical/x-xyz
Size: 9994 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210330/cb9b363f/attachment.xyz>


More information about the CP2K-user mailing list