[petsc-users] Internal memory MUMPS during numerical factorization

Matthew Knepley knepley at gmail.com
Thu Dec 9 19:53:15 CST 2021


On Thu, Dec 9, 2021 at 7:53 PM Alberto F. Martín <amartin at cimne.upc.edu>
wrote:

> Dear PETSc users,
>
> I am experiencing the following error during numerical factorization
> while using the MUMPS package.
>
> It seems an internal memory error in the DMUMPS_ALLOC_CB function. Has
> anyone experienced this before? Any clue how this can be solved?
>

I have not seen that before. It looks like an allocation failed for MUMPS,
but it is not a nice
error, and does not succumb to Google. I would mail the MUMPS developers
for this one.

  Thanks,

     Matt


> Thanks!
>
> Best regards,
>
>   Alberto.
>
> Entering DMUMPS 5.3.5 from C interface with JOB, N =   1     2359298
>        executing #MPI =      2 and #OMP =     24
>
>   =================================================
>   MUMPS compiled with option -DGEMMT_AVAILABLE
>   MUMPS compiled with option -Dmetis
>   MUMPS compiled with option -Dparmetis
>   MUMPS compiled with option -Dptscotch
>   MUMPS compiled with option -Dscotch
>   MUMPS compiled with option -DBLR_MT
>   This MUMPS version includes code for SAVE_RESTORE
>   This MUMPS version includes code for DIST_RHS
>   =================================================
> L U Solver for unsymmetric matrices
> Type of parallelism: Working host
>
>   ****** ANALYSIS STEP ********
>
>   ** Maximum transversal (ICNTL(6)) not allowed because matrix is
> distributed
> Using ParMETIS for parallel ordering.
> Structural symmetry is: 92%
>   A root of estimated size         3480  has been selected for Scalapack.
>
> Leaving analysis phase with  ...
>   INFOG(1)                                       = 0
>   INFOG(2)                                       = 0
>   -- (20) Number of entries in factors (estim.)  = 548486696
>   --  (3) Real space for factors    (estimated)  = 548486696
>   --  (4) Integer space for factors (estimated)  = 23032397
>   --  (5) Maximum frontal size      (estimated)  = 4991
>   --  (6) Number of nodes in the tree            = 173314
>   -- (32) Type of analysis effectively used      = 2
>   --  (7) Ordering option effectively used       = 2
>   ICNTL(6) Maximum transversal option            = 0
>   ICNTL(7) Pivot order option                    = 7
>   ICNTL(14) Percentage of memory relaxation      = 50000
>   Number of level 2 nodes                        = 2
>   Number of split nodes                          = 0
>   RINFOG(1) Operations during elimination (estim)= 5.292D+11
>   Distributed matrix entry format (ICNTL(18))    = 3
>
>   MEMORY ESTIMATIONS ...
>   Estimations with standard Full-Rank (FR) factorization:
>      Maximum estim. space in Mbytes, IC facto. (INFOG(16)):     1246959
>      Total space in MBytes, IC factorization (INFOG(17)):     2416245
>      Maximum estim. space in Mbytes, OOC facto. (INFOG(26)):      190319
>      Total space in MBytes,  OOC factorization (INFOG(27)):      379897
>
>   Elapsed time in analysis driver=      30.7962
>
> Entering DMUMPS 5.3.5 from C interface with JOB, N =   2 2359298
>        executing #MPI =      2 and #OMP =     24
>
> ****** FACTORIZATION STEP ********
>
> >
> >  GLOBAL STATISTICS PRIOR NUMERICAL FACTORIZATION ...
> >  Number of working processes                =               2
> >  ICNTL(22) Out-of-core option               =               0
> >  ICNTL(35) BLR activation (eff. choice)     =               0
> >  ICNTL(14) Memory relaxation                =           50000
> >  INFOG(3) Real space for factors (estimated)=       548486696
> >  INFOG(4) Integer space for factors (estim.)=        23032397
> >  Maximum frontal size (estimated)           =            4991
> >  Number of nodes in the tree                =          173314
> >  Memory allowed (MB -- 0: N/A )             =               0
> >  Memory provided by user, sum of LWK_USER   =               0
> >  Relative threshold for pivoting, CNTL(1)   =      0.1000D-01
> >  Max difference from 1 after scaling the entries for ONE-NORM (option
> > 7/8)   = 0.95D+00
> >  Average Effective size of S     (based on INFO(39))= 150930257964
> >
> >  Redistrib: total data local/sent           = 37911993        60007025
> >
> >  Elapsed time to reformat/distribute matrix =      2.3306
> >  Problem with integer stack size           1 1          13
> >  Internal error in DMUMPS_ALLOC_CB  T          15 6090000
>
>
>
>
> --
> Alberto F. Martín-Huertas
> Senior Researcher, PhD. Computational Science
> Centre Internacional de Mètodes Numèrics a l'Enginyeria (CIMNE)
> Parc Mediterrani de la Tecnologia, UPC
> Esteve Terradas 5, Building C3, Office 215,
> 08860 Castelldefels (Barcelona, Spain)
> Tel.: (+34) 9341 34223
> e-mail:amartin at cimne.upc.edu
>
> FEMPAR project co-founder
> web: http://www.fempar.org
>
> **********************
> IMPORTANT ANNOUNCEMENT
>
> The information contained in this message and / or attached file (s), sent
> from CENTRO INTERNACIONAL DE METODES NUMERICS EN ENGINYERIA-CIMNE,
> is confidential / privileged and is intended to be read only by the person
> (s) to the one (s) that is directed. Your data has been incorporated
> into the treatment system of CENTRO INTERNACIONAL DE METODES NUMERICS EN
> ENGINYERIA-CIMNE by virtue of its status as client, user of the website,
> provider and / or collaborator in order to contact you and send you
> information that may be of your interest and resolve your queries.
> You can exercise your rights of access, rectification, limitation of
> treatment, deletion, and opposition / revocation, in the terms established
> by the current regulations on data protection, directing your request to
> the postal address C / Gran Capitá, s / n Building C1 - 2nd Floor -
> Office C15 -Campus Nord - UPC 08034 Barcelona or via email to
> dpo at cimne.upc.edu
>
> If you read this message and it is not the designated recipient, or you
> have received this communication in error, we inform you that it is
> totally prohibited, and may be illegal, any disclosure, distribution or
> reproduction of this communication, and please notify us immediately.
> and return the original message to the address mentioned above.
>
>

-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener

https://www.cse.buffalo.edu/~knepley/ <http://www.cse.buffalo.edu/~knepley/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20211209/2b1d1d7a/attachment-0001.html>


More information about the petsc-users mailing list