[CPMD-list] Problem using CPMD parallel version

[Axel Kohlmeyer]

On Thu, 6 Mar 2008, oksana ismailova wrote:

OI> Dear CPMD users and experts,

dear oksana,

OI> I have problem using CPMD parallel version.
OI> 
OI>  The specs  are as follows:
OI> 
OI> Compiler: pgf90 version 5.2-4

please note that this compiler is outdated and 
known to miscompile parts of CPMD (and many other codes).
particularly the pgf90 version; pgf77 may be a little
bit more reliable. recent versions of gfortran work 
well and should be almost as fast as this one. if you
can afford it, the intel compilers will give you ~10%
speedup.

OI> RAM: 2GB
OI> CPU: AMD Opteron 248 (2.2 GHz) (AMD64)
OI> OS: SUSE 9.0
OI> Compiled using libatlas_x86_64.a from:
OI> http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/files/atlas/libatlas_
OI> x86_64.a

this library is obsolete as well, better get a recent version
of ACML from AMD. downloadable for free and with support for
gfortran and intel compiles (on top of PGI).

OI> The problem is  most of CPUs are idle.

how do you _start_ your parallel job? how does it
work for other parallel programs? this most likely
is an error on your side and not a problem with CPMD.

OI> For instance, I submit job on 10 CPUs, but 9 of them are idle, however 1 
OI> CPUs  isn't enough memory (job stop with Error indicates  have to use 
OI> more words).
OI> 
OI> Could you please give ideas to solute this problem and advices how to 
OI> reinstall CPMD parallel version.

you first have to tell us how you compile and use what you have
so we can make any suggestions on changes.

cheers,
   axel.

OI> 
OI> Thanks in advance,
OI> best regards,
OI> oksana
OI> 
OI> 
OI> 
OI> _______________________________________________
OI> CPMD-list mailing list
OI> CPMD-list at cpmd.org
OI> http://cpmd.org/mailman/listinfo/cpmd-list
OI> 

-- 
=======================================================================
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.