[CP2K-user] [CP2K:21807] Re: Cholesky decomposition failed

[Johann Pototschnig]

This problem is often caused by diffuse basis functions, so it might help 
to change:
BASIS_SET AUX_FIT cFIT3
On Monday, September 8, 2025 at 2:04:00 PM UTC+2 Johann Pototschnig wrote:

> Cholesky decomposition computes a smaller representation of a tensor, 
> which allows more efficient computations. 
>
> It assumes a positive definite matrix, which, off course, numerically can 
> not be guaranteed. 
>
> It can be switched off, but certain functionalities will not work:
>
> https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/DFT/SCF.html#CP2K_INPUT.FORCE_EVAL.DFT.SCF.CHOLESKY
>
> https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/DFT/SCF/OT.html#CP2K_INPUT.FORCE_EVAL.DFT.SCF.OT.CHOLESKY
>
> Most OT preconditioners use it. 
>
>
>
> On Monday, September 8, 2025 at 11:22:16 AM UTC+2 Dat Ho wrote:
>
>> Dear all,
>>
>>
>> I am trying to perform ab initio MD simulations on a plain water box 
>> containing 512 water molecules using the PW6B95 functional. The same input 
>> works flawlessly for smaller boxes containing 64, 128 or 265 molecules, but 
>> I can't seem to get it to work for 512 molecules. I always get the error 
>> message 'Cholesky decomposition failed'. Can anyone spot an error in the 
>> input? Also, what does this error message mean in detail? 
>>
>>
>> Any help would be much appreciated!
>>
>

-- 
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/1d201315-1e10-47d8-b136-d3826d1aa3efn%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20250908/fca189d9/attachment.htm>