[CP2K-user] [CP2K:21362] Re: performance discrepancy

[qingh...@gmail.com]

Hello Marcella, 

Thanks a lot for your reply! Now I got it. 
The cell size in nvt.inp was overwritten by the one in  EM-1.restart. 

I did see the warning, but did not realize this is due to the mismatch 
between nvt.inp and EM-1.restart,
I was only checking nvt.inp. 

Thanks again! 


All the best,
Qinghua

On Thursday, April 10, 2025 at 4:17:24 PM UTC+2 Marcella Iannuzzi wrote:

> Hi Qinghua
>
> The CELL size is different and in at lest one case there is a warning 
>
>  *** WARNING in pw_env_methods.F:742 :: The selected periodicities in the 
> ***
>  *** sections &CELL and &POISSON do not match     
>
> Regards
> Marcella
>
> On Thursday, April 10, 2025 at 3:39:52 PM UTC+2 qingh... at gmail.com wrote:
>
>> Hello everyone, 
>>
>> I performed ab initio MD simulations of alpha-glucose, beta-glucose, 
>> alpha-mannose and beta-mannose in gas phase using CP2K 2024.3 on a cluster. 
>>
>> alpha-glucose, beta-glucose, alpha-mannose and beta-mannose are isomers 
>> with 24 atoms. 
>>
>> However, the performance I got are quite different: 
>> alpha-glucose: ~6.5 s/step
>> beta-glucose: ~1.8 s/step
>> alpha-mannose: ~6.5 s/step
>> beta-mannose: ~6.5 s/step
>>
>> The difference is so big that I could not figure it out.  Could someone 
>> help me understand it?  Thanks so much! 
>>
>> I also observed this difference with CP2K 2023.1 and 2024.1.
>>
>> The attached are the input and output files. 
>>
>>
>> All the best,
>> Qinghua
>>
>>
>>

-- 
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 visit https://groups.google.com/d/msgid/cp2k/c80926b3-09a6-41b9-8351-bac6f523b864n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20250410/4c7fbce7/attachment.htm>