[CP2K-user] [CP2K:19717] Re: Changing cutoff in NVE MD run gives various errors

[Léon Luntadila Lufungula]

Hi all,

I've tested the calculation on the three different compilations on my HPC 
cluster (7.1-Intel-2020a, 2022.1-foss-2022a and 2023.2-foss-2022a) and all 
three compilations crash with FFTW_PLAN_TYPE PATIENT enabled. However, the 
Intel version does seem to throw out some different error messages. I have 
attached the input and the standard outputs of the three calculations. 

I'm planning on doing long ab intio MD calculations and as the reference 
manual states "PATIENT planning is recommended for long ab initio MD 
runs.", so I wish to exploit this feature if possible. Can someone look 
into this matter or provide an answer as to why this is happening?

Thanks in advance!

Kind regards,
Léon

On Wednesday 3 January 2024 at 16:41:13 UTC+1 Léon Luntadila Lufungula 
wrote:

> Hi everyone,
>
> So I looked into my input file a bit further and tried switching settings 
> on and off to find out if any of the settings is causing the crash and 
> apparently "FFTW_PLAN_TYPE PATIENT" is the cause of CP2K crashing with 
> CUTOFF < 800. Can someone with more knowledge of the code give their two 
> cents if this is expected behaviour or possibly a bug in the 2023.2 version 
> of CP2K?
>
> Kind regards,
> Léon
>
> On Wednesday 3 January 2024 at 14:50:31 UTC+1 Léon Luntadila Lufungula 
> wrote:
>
>> Dear all,
>>
>> A very happy 2024 to all of you!
>>
>> I amtrying to run an ab-intio NVE trajectory to check the energy 
>> conservation for a given setup of parameters (EPS_SCF, EPS_DEFAULT, CUTOFF, 
>> REL_CUTOFF, TIMESTEP EXTRAPOLATION_ORDER) to find a suitable balance 
>> between accuracy and speed. However, when trying to reduce the value for 
>> the cutoff below 800, the calculation crashes before starting the MD run 
>> with a lot of errors written to the standard output (see attached file 
>> cutoff-700.stdout). Does anyone know why this happens and if it is a 
>> compilation problem or if it is connected to my input in some way? Any help 
>> would be greatly appreciated!
>>
>> I have attached the CP2K input and output and the standard output for the 
>> calculations with CUTOFF=800 (working as expected) and CUTOFF=700 (writing 
>> errors to standard output).
>>
>> Greetings,
>> Léon
>>
>

-- 
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/05e37b95-5d96-4c61-8c61-c6d9635bb850n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240110/5682d9f1/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CP2K-7.1-intel-2020a.stdout
Type: application/octet-stream
Size: 1575 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240110/5682d9f1/attachment-0003.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CP2K-2022.1-foss-2022a.stdout
Type: application/octet-stream
Size: 24148 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240110/5682d9f1/attachment-0004.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: a101-oh-NVE-scf7-def10-patient.md.inp
Type: chemical/x-gamess-input
Size: 3270 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240110/5682d9f1/attachment-0001.inp>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CP2K-2023.2-foss-2022a.stdout
Type: application/octet-stream
Size: 7627 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240110/5682d9f1/attachment-0005.obj>