[CP2K:8668] H20-64 performance benchmark

[Ari Paavo Seitsonen]

Dear Victor,

  My guess: Just yesterday I had the same problem; the "solution" for me
was to switch to a different version, either in the compiler and/or the
Intel MPI library (I did not try all the different combinations for figure
out which one or ones were errorneous).

  'gfortran' and OpenMPI usually works much more reliably, if you cannot
try that, I would indeed recommend trying with different versions of Intel
(compilers and/or MPI).

    Greetings from Paris,

       apsi

PS The machine in question was 'Occigen' in Montpellier, the combination
that worked was

remove module impi/5.1.3.258 (intelmpi)
remove module mkl/compilers_and_libraries_2016.4.258/linux (Intel MKL)
remove module idb/compilers_and_libraries_2016.4.258/linux (Intel Debugger)
remove module intel/16.4.258 (Intel Compiler Suite)

The one that did not:

load module mkl/compilers_and_libraries_2017.0.098/linux (Intel MKL)
load module idb/compilers_and_libraries_2017.0.098/linux (Intel Debugger)
load module intel/17.0 (Intel Compiler Suite)
load module impi/2017.0.098 (intelmpi)
#> ifort --version
ifort (IFORT) 17.0.0 20160721
Copyright (C) 1985-2016 Intel Corporation.  All rights reserved.


2017-02-08 14:40 GMT+01:00 Victor K <koche... at gmail.com>:

> Hi,
>
> I'm newbie to cp2k and I'm trying to get performance numbers from cp2k
> library. According to https://www.cp2k.org/performance page one of the
> usefull benchmarks for this should be H2O-64.inp, but I'm unalbe to run it
> even for 128 ranks without some evalation error.
> I use Intel MPI, cp2k 4.1,  and run benchmark like this:
> mpirun -np 128 ../../../exe/Linux-x86-64-intel/cp2k.popt ./H2O-64.inp
>
> Could you please  advice what I'm missing while launching?
>
>
> Example output:
>
> ...
>
> ************************************************************
> *******************
> * ___ *
> * / \ *
> * [ABORT] *
> * \___/ Cholesky decompose failed: the matrix is not positive definite or *
> * | ill-conditioned. *
> * O/| *
> * /| | *
> * / \ fm/cp_fm_cholesky.F:94 *
> ************************************************************
> *******************
>
>
> ===== Routine Calling Stack =====
>
> 11 cp_fm_cholesky_decompose
> 10 make_basis_sm
> 9 reorthogonalize_vectors
> 8 wfi_extrapolate
> 7 scf_env_initial_rho_setup
> 6 init_scf_run
> 5 qs_energies
> 4 qs_forces
> 3 velocity_verlet
> 2 qs_mol_dyn_low
> 1 CP2K
>
> --
> You received this message because you are subscribed to the Google Groups
> "cp2k" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to cp2k+uns... at googlegroups.com.
> To post to this group, send email to cp... at googlegroups.com.
> Visit this group at https://groups.google.com/group/cp2k.
> For more options, visit https://groups.google.com/d/optout.
>



-- 
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
  Ari Paavo Seitsonen / Ari.P.S... at iki.fi / http://www.iki.fi/~apsi/
    Ecole Normale Supérieure (ENS), Département de Chimie, Paris
    Mobile (F) : +33 789 37 24 25    (CH) : +41 79 71 90 935
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170208/9d038de1/attachment.htm>