[CP2K:3639] how to run a parallel work in cp2k

[Jörg Saßmannshausen]

Dear Zhuhong

you need to run it that way:

mpirun -np 4 /path/to/cp2k.popt INPUTFILE > OUTPUTFILE

There is no < after the cp2k.popt, i.e. no redirection.

Also, I think you don't need the libint-wrapper anymore?

All the best from a sunny London :D

Jörg


On Friday 25 November 2011 11:33:33 zhuhong zhu wrote:
> dear all:
> I'm a beginner of CP2K and I want to compile a parallel version of
> cp2k (cp2k.popt).
> After i modified the Linux-x86-64-intel.popt and compiled it
> successfully, I really got a file named cp2k.popt. but when I try to
> run mpirun -np 4 cp2k.popt <input_file>. I found four individual
> cp2k.popt runing. four same result are output to the terminal. So who
> can tell me the details of compiling and running the parallel version
> of cp2k.
> 
> below is the   Linux-x86-64-intel.popt file I used for compile.
> 
> INTEL_INC=/usr/local/include
> INTEL_LIB=/opt/intel/mkl/10.0.5.025/lib/em64t/
> CC       = cc
> CPP      =
> FC       = mpif90
> LD       = mpif90
> AR       = ar -r
> DFLAGS   = -D__INTEL -D__FFTSG -D__parallel -D__BLACS -D__SCALAPACK -
> D__FFTW3 -D__LIBINT
> CPPFLAGS =
> FCFLAGS  = $(DFLAGS) -I$(INTEL_INC) -O3 -xW -heap-arrays 64 -funroll-
> loops -fpp -free
> FCFLAGS2 = $(DFLAGS) -I$(INTEL_INC) -O1 -xW -heap-arrays 64 -fpp -free
> LDFLAGS  = $(FCFLAGS) -I$(INTEL_INC)
> LIBS     =  -L$(INTEL_LIB) -lmkl_intel_lp64 -lmkl_sequential -lmkl_core
> \
>             -lmkl_scalapack_lp64 -lmkl_blacs_intelmpi_lp64 \
>             -lmkl_blacs_openmpi_lp64 \
>         -lmkl_intel_lp64 -lmkl_sequential -lmkl_core  -lpthread \
>            -lfftw3 \
>            /home/zhuh/cp2k/tools/hfx_tools/libint_tools/
> libint_cpp_wrapper.o\
>            /usr/local/libint/1.1.4-stable/lib/libderiv.a\
>            /usr/local/libint/1.1.4-stable/lib/libint.a\
>            -lstdc++
> 
> OBJECTS_ARCHITECTURE = machine_intel.o
> 
> 
> graphcon.o: graphcon.F
>         $(FC) -c $(FCFLAGS2) $<
> 
> best wishes for all from hefei a small cold city.

-- 
*************************************************************
Jörg Saßmannshausen
University College London
Department of Chemistry
Gordon Street
London
WC1H 0AJ 

email: j.sassma... at ucl.ac.uk
web: http://sassy.formativ.net

Please avoid sending me Word or PowerPoint attachments.
See http://www.gnu.org/philosophy/no-word-attachments.html