[Wien] LAPW2 crashed on Altix 3000 but not on Dell cluster

[Laurence Marks]

Almost certainly different numerical accuracy and/or treatment of
rounding errors and other things. All fortran compilers are equal, but
some are more equal than others. The same for linux/unix
implimentations.

On 3/20/07, Shen Li Qiu <qiu at fau.edu> wrote:
> Hi Laurence,
>
> At 04:02 PM 3/19/2007 -0500, you wrote:
> >Look at case.output2 to try and see why it crashed. Probably you were
> >using GAUSS smearing which can be flakey (numerically) -- TEMP is more
> >robust. Alternatively you did not go high enough in energy in
> >case.in1, or had too many eigenvalues.
> >
>
> My question is: why the same job does not crash on Dell cluster but crashes
> on Altix 3000? The same version of WIEN2k_07 and same compiler are used on
> both machines and all the parameters used in the initializations are the same.
>
> Shen Li
>
>
> >On 3/19/07, Shen Li Qiu <qiu at fau.edu> wrote:
> >> Hi all,
> >> LAPW2 crashed on Altix 3000 with the following messages:
> >>
> >> (1) Message in the "dayfile":
> >> >   lapw2 -c -up -p     (10:19:08) running LAPW2 in parallel mode
> >> **  LAPW2 crashed!
> >> 1.637u 0.191s 0:03.10 58.7%     0+0k 0+0io 0pf+0w
> >> error: command   /home/qiu/WIEN07/lapw2cpara -up -c uplapw2.def   failed
> >> >   stop error
> >>
> >> (2) Message in "STDOUT":
> >> LAPW1 END
> >> FERMI - Error
> >> cp: cannot stat `.in.tmp': No such file or directory
> >> rm: cannot remove `.in.tmp': No such file or directory
> >> rm: cannot remove `.in.tmp1': No such file or directory
> >> "STDOUT" 25L, 409C
> >>
> >>  The latest version of WIEN2k_07 was compiled without error on Altix 3000
> >> (with shared memory and MPI) using ifort and icc.
> >>
> >>  Exactly the same calculation runs OK on a Dell cluster (without shared
> >> memory and MPI) using the latest version of WIEN2k_07, which indicates that
> >> the crash of lapw2 on Altix 3000 should not be due to the calculation
> itself.
> >>
> >>  The compiling settings on Altix 3000 are:
> >>  Current settings:
> >>      O   Compiler options:        -O3 -tpp2 -ftz -ip -fno-alias -cm -w -FR
> >> -DALTIX -DINTEL_VML
> >>      L   Linker Flags:            -L../SRC_lib -Vaxlib
> >>      P   Preprocessor flags       '-DParallel'
> >>      R   R_LIB (LAPACK+BLAS):     -L/usr/local/GotoBLAS1.06
> >> -lgoto_ITANIUM2p-r1.06 -lscs -L/opt/intel/mkl/8.0.2/lib/64 -lvml
> >>
> >> Please help me to figure out what causes the crash of lapw2 on Altix 3000.
> >> Please let me know if you need further information.
> >>
> >> Thanks in advance.
> >>
> >> Shen Li Qiu
> >>
> >>
> >>
> >> _______________________________________________
> >> Wien mailing list
> >> Wien at zeus.theochem.tuwien.ac.at
> >> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> >>
> >
> >
> >--
> >Laurence Marks
> >Department of Materials Science and Engineering
> >MSE Rm 2036 Cook Hall
> >2220 N Campus Drive
> >Northwestern University
> >Evanston, IL 60208, USA
> >Tel: (847) 491-3996 Fax: (847) 491-7820
> >email: L-marks at northwestern dot edu
> >Web: www.numis.northwestern.edu
> >EMM2007 http://ns.crys.ras.ru/EMMM07/
> >_______________________________________________
> >Wien mailing list
> >Wien at zeus.theochem.tuwien.ac.at
> >http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> >
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>


-- 
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60208, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
Web: www.numis.northwestern.edu
EMM2007 http://ns.crys.ras.ru/EMMM07/