[Wien] lapw2 stops

[Oleg Artamonov]

Dear Peter Blaha,

Thanks for the quick replay.

> Hard to guess!
>
> Maybe NUME is not big enough, so you don't have enough eigenvalues to
> fill in all electrons ?

NOE=374 for 22-layer surface structure and NUME=1000 in the param.inc
>
> Eventually you do not have enough memory ?

RAM=4GB and less than 1GB is in use
>
> Try to insert a line like:
>
> lapw2_vector_split:2   (or 4)
>
> into .machines
>
> It will allocate less memory in lapw2_mpi
>
I inserted lapw2_vector_split:2 as well as lapw2_vector_split:4,
but without success: lapw2 stops " Segmentation fault".

Thank you,
                    Oleg Artamonov/


> Oleg Artamonov schrieb:
>> Dear Wien users and leaders,
>>
>> The run_lapw (as well as run_lapw -p) works normal for the 7-layer 
>> surface
>> of W and stops at the beginning of lapw2 for
>> the 22-layer surface of Au.
>>
>> We use (48 node *4 core) cluster with 2*Intel Xeon 5335, OS = Linux SLES 
>> 10
>> SP1,
>> for parallelization - mvapich-1/0/0-1844, infinityband.
>> The siteconfig options are:
>> 1) For MPI ( mvapich ) parallelization:
>>
>> RP_LIB(SCALAPACK+PBLAS): -lmkl_intel_lp64 -lmkl_scalapack_lp64
>> -lmkl_blacs_lp64 -lmkl_sequential -lmkl_em64t
>>
>> FPOPT(par.comp.options): $(FOPT) -lmkl_intel_lp64 -lmkl_scalapack_lp64
>> -lmkl_blacs_lp64 -lmkl_sequential -lmkl_core -liomp5 -pthread -i-static
>>
>> MPIRUN commando        : /share/mvapich-1.0.1-intel/bin/mpirun_rsh
>>
>> 2) For serial work:
>> Compiler 
>> options:        -FR -mp1 -prec_div -r8 -pc80 -fpconstant -traceback
>> -pad -align -O3 -ipo1 -DINTEL_VML -i-static -fminshared -xT -thread
>> -DINTEL_VML
>>
>> Linker Flags:            -L/opt/intel/mkl/10.0.3.020/lib/em64t
>>
>> Preprocessor flags       '-DParallel'
>>
>> R_LIB (LAPACK+BLAS):     $(FPOPT) -lmkl_intel_lp64 -lmkl_scalapack_lp64
>> -lmkl_blacs_lp64 -lmkl_sequential -lmkl_core -liomp5 -pthread -i-static
>>
>> Thanks in advance for any suggestions,
>> 
>> Oleg
>> Artamonov/
>>
>>
>>
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
> -- 
>
>                                       P.Blaha
> --------------------------------------------------------------------------
> Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
> Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
> Email: blaha at theochem.tuwien.ac.at    WWW: 
> http://info.tuwien.ac.at/theochem/
> --------------------------------------------------------------------------
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>