[CP2K-user] Help: The setting of input file for calculating absorption spectra of large system by cp2k

[lin xiong]

Hi everyone,

     Recently I wanted to calculate the UV-Vis of some large systems, so I 
started to learn cp2k. But some troubles were encountered in this process. 
For example, if the spectrum is calculated on a node with a running memory 
of 250G, the remaining memory is quite tight when proceeding to the "TDDFPT 
WAVEFUNCTION OPTIMIZATION" step. Due to the huge size of my system (with 
small HOMO-LUMO gap), I had to set more nstates in the inp file. In 
addition to calculating more excited states, does the inp file have some 
settings that can be optimized? I look forward to and thank you for your 
advice.

     Remarks: The CPU version information is 96 Intel(R) Xeon(R) Platinum 
8175M CPU @ 2.50GHz. The disk space is 4T.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200705/720a184f/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 246pbe.inp
Type: chemical/x-gamess-input
Size: 6322 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200705/720a184f/attachment.inp>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: out
Type: application/octet-stream
Size: 33875 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200705/720a184f/attachment.obj>