[CP2K-user] [CP2K:16246] Re: The SCF iteration rate in QM/MM is significantly different from that in vacuo

[Marcella Iannuzzi]

Hi

The QM box in your QMMM calculation is much larger. 
A considerable amount of time is going into PW and FFT operations and 
calculation of QMMM forces
 pw_transfer                       1450 11.7    0.271    0.334  157.786 
 161.976
 fft_wrap_pw1pw2                   1282 12.8    0.028    0.036  155.109 
 159.276
 fft_wrap_pw1pw2_150                822 14.0   17.073   19.058  151.008 
 154.414
qmmm_forces                          4  3.8    0.001    0.003  117.781 
 117.782
 qmmm_forces_with_gaussian            4  4.8    0.088    0.098  115.457 
 116.497
 fft3d_ps                          1282 14.8   63.372   66.540  107.554 
 110.807
 pw_restrict_s3                      12  5.8   31.639   32.981  101.177 
 102.309
 qmmm_el_coupling                     4  3.8    0.000    0.000   97.206   
98.164
 qmmm_elec_with_gaussian              4  4.8    0.062    0.104   97.185   
98.142

Regards
Marcella

On Wednesday, November 17, 2021 at 3:14:20 AM UTC+1 bnuyegu... at gmail.com 
wrote:

> Dear all, 
>
>       I have tested the molecular dynamics of the same molecule in the gas 
> phase and in the QM/MM environment, and found that the SCF iteration of 
> QM/MM is dozens of times slower than that of vacuum.  Is this 
> reasonable? As far as I know, the coupling between QM and MM should not 
> affect the speed so much. 
>
> The input and output files for the tests are  attached.
>
>
>
>
>

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/1b67e739-d69f-46d9-aaaa-15007b4ac8ban%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20211116/79da8d72/attachment.htm>