[CPMD-list] BO-MD starting problem after WFO (at different nubmer of CPUs)

[Axel Kohlmeyer]

On Tue, 12 Jun 2007, Vladimir Stegailov wrote:

VS> Dear colleagues,
VS> 
VS> recently I had a problem that BO-MD runs starting from an optimized WF
VS> should start from optimization process once again. The reason was that I
VS> stored WF with lossy compression (COMPRESS WRITE32).
VS> 
VS> However now I have got the similar problem again. If the WF optimization
VS> and the subsequent MD run are executed using different number of cpus
VS> (e.g. 8 or 6) – the force initialization before BO-MD tends to fail. As if
VS> the WF has been not optimized at all. 
VS> 
VS> What could be the reason for that?

vladimir,

restart files do not encode the number of CPUs in any way,
so the problem is most likely either in your machine setup, 
your MPI library or your compiler.

cheers,
   axel.

VS> 
VS> Vladimir
VS> 
VS> 
VS> --===============1214697718==
VS> Content-Type: text/plain; charset="us-ascii"
VS> MIME-Version: 1.0
VS> Content-Transfer-Encoding: 7bit
VS> Content-Disposition: inline
VS> 
VS> _______________________________________________
VS> CPMD-list mailing list
VS> CPMD-list at cpmd.org
VS> http://cpmd.org/mailman/listinfo/cpmd-list
VS> 
VS> --===============1214697718==--
VS> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.