search

FreeFem / FreeFem-sources

FreeFEM source code
https://freefem.org/
Other
770 stars 191 forks source link

Problems with ffddm in FreeFem++ #103

Closed zhaog6 closed 5 years ago

zhaog6 commented 5 years ago

Dear all,

I have a problem with ffddm. I downloaded FreeFEM(v4.2-1) from github (https://github.com/FreeFem/FreeFem-sources.git) and installed on Ubuntu 18.04 correctly according to the installation guide. When running the file diffusion-3d-simple.edp in ffddm, I met the following compiling and running error (load error: metis, and undefined symbol: METIS_PartMeshNodal in the file metis.so).

(But It works well using FreeFem++-4.2-1-b-348-MACOS_10.11.pkg on Mac)

   15 : IFMACRO(dimension,3S)
   16 & load "msh3"
   17 & macro meshN()meshS// EOM            // three-dimensional surface problem
macro intN()int2d// EOM
macro intbN()int1d// EOM
macro minimalMesh()square3(1,1,[x,y,z])// EOM
ENDIFMACRO
   18 : 
   19 : IFMACRO(WITHHPDDM)
   20 & load "hpddm"
   21 & ENDIFMACRO
  current line = 2 mpirank 2 / 4
   20 : load "hpddm"
   21 : 
   22 : 
  current line = 2 mpirank 3 / 4
   23 : load "thresholdings"
   24 : 
  current line = 2 mpirank 1 / 4
   25 : include "ffddm_partitioning.idp"load "scotch"
    2 : load "metis"
load error : metis
     fail : 
 dlerror : /home/zhaog6/zg/FreeFEM/4.2-1/lib/ff++/4.2-1/lib/mpi/metis.so: undefined symbol: METIS_PartMeshNodal
list  prefix: './' '/home/zhaog6/zg/FreeFEM/4.2-1/lib/ff++/4.2-1/lib/mpi/' list  suffix : '' , '.so' 
  current line = 2 mpirank 0 / 4
Load error : metis
    line number :2, metis
error Load error : metis
    line number :2, metis
 code = 2 mpirank: 0
-------------------------------------------------------
Primary job  terminated normally, but 1 process returned
a non-zero exit code.. Per user-direction, the job has been aborted.
-------------------------------------------------------
--------------------------------------------------------------------------
mpirun detected that one or more processes exited with non-zero status, thus causing
the job to be terminated. The first process to do so was:

  Process name: [[33165,1],0]
  Exit code:    2
--------------------------------------------------------------------------
frederichecht commented 5 years ago

Dear The problem is due to the plugin metis,
but I down know why it is not compile. I will see soon has possible but I am I holiday,

Best Regards,

Frédéric Hecht.

Laboratoire Jacques-Louis Lions, UPMC Sorbonne Université BC187, 4 Place Jussieu, 75252 PARIS cedex 05, France Campus Jussieu, Barre 15-25, 3 etage Bureau 307 Projet Alpines , Inria de Paris, 2 rue Simone Iff Voie DQ12 75012 Paris tel: +33 1 44274411, mob: +33 6 62198986, fax: +33 1 44277200 frederic.hecht@sorbonne-universite.fr https://www.ljll.math.upmc.fr/hecht software: FreeFem++ web site: http://www.freefem.org/ff++

Le 17 juil. 2019 à 07:03, zhaog6 notifications@github.com a écrit :

Dear all,

I have a problem with ffddm. I downloaded FreeFEM(v4.2-1) from github (https://github.com/FreeFem/FreeFem-sources.git) and installed on Ubuntu 18.04 correctly according to the installation guide. When running the file diffusion-3d-simple.edp in ffddm, I met the following compiling and running error (load error: metis, and undefined symbol: METIS_PartMeshNodal in the file metis.so).

(But It works well using FreeFem++-4.2-1-b-348-MACOS_10.11.pkg on Mac)

15 : IFMACRO(dimension,3S) 16 & load "msh3" 17 & macro meshN()meshS// EOM // three-dimensional surface problem macro intN()int2d// EOM macro intbN()int1d// EOM macro minimalMesh()square3(1,1,[x,y,z])// EOM ENDIFMACRO 18 : 19 : IFMACRO(WITHHPDDM) 20 & load "hpddm" 21 & ENDIFMACRO current line = 2 mpirank 2 / 4 20 : load "hpddm" 21 : 22 : current line = 2 mpirank 3 / 4 23 : load "thresholdings" 24 : current line = 2 mpirank 1 / 4 25 : include "ffddm_partitioning.idp"load "scotch" 2 : load "metis" load error : metis fail : dlerror : /home/zhaog6/zg/FreeFEM/4.2-1/lib/ff++/4.2-1/lib/mpi/metis.so: undefined symbol: METIS_PartMeshNodal list prefix: './' '/home/zhaog6/zg/FreeFEM/4.2-1/lib/ff++/4.2-1/lib/mpi/' list suffix : '' , '.so' current line = 2 mpirank 0 / 4 Load error : metis line number :2, metis error Load error : metis line number :2, metis code = 2 mpirank: 0

Primary job terminated normally, but 1 process returned a non-zero exit code.. Per user-direction, the job has been aborted.

mpirun detected that one or more processes exited with non-zero status, thus causing the job to be terminated. The first process to do so was:

Process name: [[33165,1],0] Exit code: 2

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/FreeFem/FreeFem-sources/issues/103?email_source=notifications&email_token=ABLPNDG4GEAMMY35MV4Z6FDP72R27A5CNFSM4IEL2VFKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4G7UFWXQ, or mute the thread https://github.com/notifications/unsubscribe-auth/ABLPNDFRHWAD733V6JZUSRLP72R27ANCNFSM4IEL2VFA.

prj- commented 5 years ago

This should be fixed now with the latest revision of the develop branch. About your question here, you can use ffddm (folder examples/ffddm, documentation) or HPDDM and PETSc (folder examples/hpddm, documentation). The largest problem I solved with FreeFEM + PETSc had around 50 billion unknowns. Keep in mind that PETSc may scale up to trillion of unknowns, so there is no real limit.

zhaog6 commented 5 years ago

The plugin “ffddm” in the newest version on the develop branch works well on my server now.Thank you for your help!

在 2019年8月23日，01:40，prj- notifications@github.com 写道：

This should be fixed now with the latest revision of the develop branch. About your question here https://community.freefem.org/t/a-problem-of-parallel-computation-in-3d/95, you can use ffddm (folder examples/ffddm, documentation https://doc.freefem.org/documentation/ffddm/documentation.html) or HPDDM and PETSc (folder examples/hpddm, documentation http://jolivet.perso.enseeiht.fr/FreeFem-tutorial/). The largest problem I solved with FreeFEM + PETSc had around 50 billion unknowns. Keep in mind that PETSc may scale up to trillion of unknowns, so there is no real limit.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/FreeFem/FreeFem-sources/issues/103?email_source=notifications&email_token=AHT7HSVLCHIMQXNHDVYB4WLQF3FQZA5CNFSM4IEL2VFKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD453AQI#issuecomment-524005441, or mute the thread https://github.com/notifications/unsubscribe-auth/AHT7HSRNTL2HMZGZJZYPGNDQF3FQZANCNFSM4IEL2VFA.