[CP2K-user] [CP2K:19618] Time of Diagonalization/OT steps increases with simulation time

[Krack Matthias]

Dear Benedikt

The system does not converge starting from about step 30 for a while and during that time the timings for an SCF iteration increase gradually. No clue, why that happens and I am even surprised that the system recovers from that “instability” period without going bananas completely. Btw, you should disable the OUTER_SCF with Broyden mixing, since that makes only sense with OT. Why you are using the CP2K version 6.1 from 2018? Many issues have been fixed in the meanwhile. It is not very likely that somebody will have a look at that old version. Moreover, such problems with inhomogeneous timings are often caused on the system side rather than on the CP2K side.

Best

Matthias

From: cp2k at googlegroups.com <cp2k at googlegroups.com> on behalf of Benedikt Schier <benedikt.schier at gmail.com>
Date: Friday, 1 December 2023 at 11:55
To: cp2k <cp2k at googlegroups.com>
Subject: Re: [CP2K:19616] Time of Diagonalization/OT steps increases with simulation time
Dear Matthias,
Actually the number of SCF steps is fine, this number does not increase significantly with the simulation time. I am talking about the time per iteration step (sorry for not explaining it sufficiently) of the diagonalisation/OT step. I have attached a screenshot to make it clearer.The output file seems to be too large, here is a google drive link: https://drive.google.com/drive/folders/1COlF3zyv4nGsQ5e0sTnnG82oBm2cwPf8?usp=sharing

With kind regards,

Benedikt Schier
--
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<mailto:cp2k+unsubscribe at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/9e65b09e-f766-4dfe-87d1-c92dcc1342e1n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/9e65b09e-f766-4dfe-87d1-c92dcc1342e1n%40googlegroups.com?utm_medium=email&utm_source=footer>.

-- 
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/ZRAP278MB08271D578B57F7D5FC06ACF6F481A%40ZRAP278MB0827.CHEP278.PROD.OUTLOOK.COM.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20231201/da41f1c5/attachment.htm>