[CP2K-user] [CP2K:19930] Geometric optimization hangs with no error.

[Krack Matthias]

Hi

You can also set the keywords TRACE<https://manual.cp2k.org/trunk/CP2K_INPUT/GLOBAL.html#CP2K_INPUT.GLOBAL.TRACE> or only TRACE_MASTER<https://manual.cp2k.org/trunk/CP2K_INPUT/GLOBAL.html#CP2K_INPUT.GLOBAL.TRACE_MASTER> in the section &GOBAL to get more information printed during a run.

HTH

Matthias

From: cp2k at googlegroups.com <cp2k at googlegroups.com> on behalf of Max Nishibuchi <max at quantumventura.com>
Date: Friday, 16 February 2024 at 19:46
To: cp2k <cp2k at googlegroups.com>
Subject: Re: [CP2K:19921] Geometric optimization hangs with no error.
Thank you very much Matthias, I was able to get the calculation to converge using one of the recent CPU docker images.

However, I'm still not entirely sure how to go about diagnosing the issue with the GPU image. Is PRINT_LEVEL the only way to adjust how much information is written/output by the program?
Kind regards,

Max
On Wednesday, February 14, 2024 at 12:12:57 PM UTC-5 Krack Matthias wrote:
Hi

Firstly, I would make sure that your test input runs smoothly on a standard CPU (cluster) system.
Note, that there are CP2K containers<https://github.com/cp2k/cp2k-containers> available for the latest release v2024.1 including containers for P100 on docker hub<https://hub.docker.com/r/cp2k/cp2k/tags/>.

Matthias

From: cp... at googlegroups.com <cp... at googlegroups.com> on behalf of Max Nishibuchi <m... at quantumventura.com>
Date: Wednesday, 14 February 2024 at 17:19
To: cp2k <cp... at googlegroups.com>
Subject: [CP2K:19903] Geometric optimization hangs with no error.
Hi guys,

I'm new to CP2K but not new to DFT methods in general. I have been trying to perform a basic geometry optimization on a styrene molecule and am finding that the simulation will run for several optimization steps, but it never finishes.

The geometry optimization will simply hang in the middle of the optimization with no input error provided regardless of the PRINT_LEVEL parameter I set. I've included the input file here. If it makes any difference, I am running this using the CUDA docker image (v2023.1) on an H100. I don't believe that this is an OOM error, as memory usage during the optimization on both the GPU and CPU is far, far below their maximum capacities.

Is there anything I can use besides PRINT_LEVEL to diagnose what is causing my simulations to hang?
--
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+uns... at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/6c8babd0-bb50-424d-8a40-24fe8660ef09n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/6c8babd0-bb50-424d-8a40-24fe8660ef09n%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<mailto:cp2k+unsubscribe at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/2d71b7b9-af3e-432a-adf0-88980799d303n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/2d71b7b9-af3e-432a-adf0-88980799d303n%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/ZRAP278MB08276DFEA0A2705281DAB1CFF4512%40ZRAP278MB0827.CHEP278.PROD.OUTLOOK.COM.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240219/b5fe57e6/attachment-0001.htm>