changes between 2.1 and 2.3/2.4-trunk inside calculate_dispersion_pairpot subroutine

[Marcin Zielinski]

Hi,

I've recently came across a problem with the 'calculate_dispersion_pairpot' subroutine inside the qs_dispersion_pairpot.F file. The 2.4 version of that subroutine is being executed ~6x slower than that of 2.1 (~6s in place of ~1s). This was initially reported to me by one of our users running his jobs on our cluster. The builds of both 2.4 and 2.1 are practically the same.

Any thoughts and ideas about it?

regards and thanks in advance,

[hut...@pci.uzh.ch]

Hi

I cannot reproduce this behaviour. Can you send an example and
specify the job parameters (serial/parallel, number of nodes)?

thank you

Juerg

--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Physical Chemistry Institute FAX : ++41 44 635 6838
University of Zurich E-mail: hut...@pci.uzh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: cp...@googlegroups.com
From: Marcin Zielinski
Sent by: cp...@googlegroups.com
Date: 05/01/2013 12:26PM
Subject: [CP2K:4408] changes between 2.1 and 2.3/2.4-trunk inside calculate_dispersion_pairpot subroutine

--
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...@googlegroups.com.
To post to this group, send email to cp...@googlegroups.com.
Visit this group at http://groups.google.com/group/cp2k?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Marcin Zielinski]

Hi,

I've asked the research group in question to send you the test case and also to try to find out what could be happening there. 

I on my side, can provide the configuration files for both version for sanity check and perhaps the profiling results. 

I'll email those to you shortly.

best regards and thanks in advance,

Op vrijdag 3 mei 2013 12:48:09 UTC+2 schreef jgh het volgende: