Re: [Wien] Error in mpi+k point parallelization across multiple nodes
Ok, now it is clear that there is no additional error messages. Unfortunately, I cannot tell specifically what went wrong from those error messages. You might try replacing mpirun with mpirun_rsh. As you can see at http://mailman.cse.ohio-state.edu/pipermail/mvapich-discuss/2013-May/004402.html they replaced mpirun with mpirun_rsh, and it seems that they found a problem with passwordless ssh. In your parallel_options file, you might also want to change setenv USE_REMOTE 0 to setenv USE_REMOTE 1, then try both 0 and 1 for MPI_REMOTE to check if any of these other configurations work or not while using mpirun. On 5/6/2015 11:23 AM, lung Fermin wrote: Thanks for the reply. Please see below. As I asked before, did you give us all the error information in the case.dayfile and from standard output? It is not entirely clear in your previous posts, but it looks to me that you might have only provided information from the case.dayfile and the error files (cat *.error), but maybe not from the standard output. Are you still using the PBS script in your old post at http://www.mail-archive.com/wien%40zeus.theochem.tuwien.ac.at/msg11770.html ? In the script, I can see that the standard output is set to be written to a file called wien2k_output. Sorry for the confusion. Yes, I still use the PBS script in the above link. The posts before are from the standard outputs (wien2k). When using 2 nodes with 32 cores for one k point, the standard output gives Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-1 8 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 number of processors: 32 LAPW0 END [16] Failed to dealloc pd (Device or resource busy) [0] Failed to dealloc pd (Device or resource busy) [17] Failed to dealloc pd (Device or resource busy) [2] Failed to dealloc pd (Device or resource busy) [18] Failed to dealloc pd (Device or resource busy) [1] Failed to dealloc pd (Device or resource busy) LAPW1 END LAPW2 - FERMI; weighs written [z1-17:mpispawn_0][child_handler] MPI process (rank: 0, pid: 28291) terminated with signal 9 - abort job [z1-17:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 9. MPI process died? [z1-17:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-17:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-17 aborted: Error while reading a PMI socket (4) [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][handle_mt_peer] Error while reading PMI socket. MPI process died? cp: cannot stat `.in.tmp': No such file or directory stop error - And the .dayfile reads: on z1-17 with PID 29439 using WIEN2k_14.2 (Release 15/10/2014) start (Thu Apr 30 17:36:59 2015) with lapw0 (40/99 to go) cycle 1 (Thu Apr 30 17:36:59 2015) (40/99 to go) lapw0 -p (17:36:59) starting parallel lapw0 at Thu Apr 30 17:36:59 2015 .machine0 : 32 processors 904.074u 8.710s 1:01.54 1483.2% 0+0k 239608+78400io 105pf+0w lapw1 -p -c (17:38:01) starting parallel lapw1 at Thu Apr 30 17:38:01 2015 - starting parallel LAPW1 jobs at Thu Apr 30 17:38:01 2015 running LAPW1 in parallel mode (using .machines) 1 number_of_parallel_jobs z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18(8) 469689.261u 1680.003s 8:12:29.52 1595.1% 0+0k 204560+31265944io 366pf+0w Summary of lapw1para: z1-17 k=0 user=0 wallclock=0 469788.683u 1726.356s 8:12:31.33 1595.5%0+0k 206128+31266512io 379pf+0w lapw2 -p-c (01:50:32) running LAPW2 in parallel mode z1-17 0.034u 0.040s 1:35.16 0.0% 0+0k 10696+0io 80pf+0w Summary of lapw2para: z1-17 user=0.034 wallclock=95.16 ** LAPW2 crashed! 4.645u 0.458s 1:42.01 4.9% 0+0k 74792+45008io 133pf+0w error: command /home/stretch/flung/DFT/WIEN2k/lapw2cpara -c lapw2.def failed stop error - When it runs fine on a single node, does it always use the same node (say z1-17) or does it run fine on other nodes (like z1-18)? Not really. The nodes were assigned randomly. ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
Thanks for the reply. Please see below. As I asked before, did you give us all the error information in the case.dayfile and from standard output? It is not entirely clear in your previous posts, but it looks to me that you might have only provided information from the case.dayfile and the error files (cat *.error), but maybe not from the standard output. Are you still using the PBS script in your old post at http://www.mail-archive.com/wien%40zeus.theochem.tuwien.ac.at/msg11770.html ? In the script, I can see that the standard output is set to be written to a file called wien2k_output. Sorry for the confusion. Yes, I still use the PBS script in the above link. The posts before are from the standard outputs (wien2k). When using 2 nodes with 32 cores for one k point, the standard output gives Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-1 8 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 number of processors: 32 LAPW0 END [16] Failed to dealloc pd (Device or resource busy) [0] Failed to dealloc pd (Device or resource busy) [17] Failed to dealloc pd (Device or resource busy) [2] Failed to dealloc pd (Device or resource busy) [18] Failed to dealloc pd (Device or resource busy) [1] Failed to dealloc pd (Device or resource busy) LAPW1 END LAPW2 - FERMI; weighs written [z1-17:mpispawn_0][child_handler] MPI process (rank: 0, pid: 28291) terminated with signal 9 - abort job [z1-17:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 9. MPI process died? [z1-17:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-17:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-17 aborted: Error while reading a PMI socket (4) [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][handle_mt_peer] Error while reading PMI socket. MPI process died? cp: cannot stat `.in.tmp': No such file or directory stop error - And the .dayfile reads: on z1-17 with PID 29439 using WIEN2k_14.2 (Release 15/10/2014) start (Thu Apr 30 17:36:59 2015) with lapw0 (40/99 to go) cycle 1 (Thu Apr 30 17:36:59 2015) (40/99 to go) lapw0 -p(17:36:59) starting parallel lapw0 at Thu Apr 3017:36:59 2015 .machine0 : 32 processors 904.074u 8.710s 1:01.54 1483.2% 0+0k 239608+78400io 105pf+0w lapw1 -p -c (17:38:01) starting parallel lapw1 at Thu Apr 30 17:38:01 2015 - starting parallel LAPW1 jobs at Thu Apr 30 17:38:01 2015 running LAPW1 in parallel mode (using .machines) 1 number_of_parallel_jobs z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18(8) 469689.261u 1680.003s 8:12:29.52 1595.1% 0+0k 204560+31265944io 366pf+0w Summary of lapw1para: z1-17 k=0 user=0 wallclock=0 469788.683u 1726.356s 8:12:31.33 1595.5%0+0k 206128+31266512io 379pf+0w lapw2 -p -c (01:50:32) running LAPW2 in parallel mode z1-17 0.034u 0.040s 1:35.16 0.0% 0+0k 10696+0io 80pf+0w Summary of lapw2para: z1-17 user=0.034 wallclock=95.16 ** LAPW2 crashed! 4.645u 0.458s 1:42.01 4.9% 0+0k 74792+45008io 133pf+0w error: command /home/stretch/flung/DFT/WIEN2k/lapw2cpara -c lapw2.def failed stop error - When it runs fine on a single node, does it always use the same node (say z1-17) or does it run fine on other nodes (like z1-18)? Not really. The nodes were assigned randomly. ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
See below for my comments. Thanks for all the information and suggestions. I have tried to change -lmkl_blacs_intelmpi_lp64 to -lmkl_blacs_lp64 and recompile. However, I got the following error message in the screen output LAPW0 END [cli_14]: [cli_15]: [cli_6]: aborting job: Fatal error in PMPI_Comm_size: Invalid communicator, error stack: PMPI_Comm_size(110): MPI_Comm_size(comm=0x5b, size=0x7f190c) failed PMPI_Comm_size(69).: Invalid communicator aborting job: Fatal error in PMPI_Comm_size: Invalid communicator, error stack: PMPI_Comm_size(110): MPI_Comm_size(comm=0x5b, size=0x7f190c) failed PMPI_Comm_size(69).: Invalid communicator ... [z0-5:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 20. MPI process died? [z0-5:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z0-5:mpispawn_0][child_handler] MPI process (rank: 14, pid: 11260) exited with status 1 [z0-5:mpispawn_0][child_handler] MPI process (rank: 3, pid: 11249) exited with status 1 [z0-5:mpispawn_0][child_handler] MPI process (rank: 6, pid: 11252) exited with status 1 . This is probably because you are using the wrong blacs library. The -lmkl_blacs_lp64 is for MPICH, but you are using a variant of MPICH3. Previously I compiled the program with -lmkl_blacs_intelmpi_lp64 and the mpi parallelization on a single node seems to be working. I notice that during the run, the *.error files have finite sizes, but I re-examine them after the job finished and there were no errors written inside (and the files have 0kb now). Does this indicates that the mpi is not running probably at all even on a single node? But I have checked the output result and it's in agreement with the non-mpi results..(for some simple cases) Sounds like it is working fine on a single node. At least for now, stay with -lmkl_blacs_intelmpi_lp64 as it works for a single node. As I asked before, did you give us all the error information in the case.dayfile and from standard output? It is not entirely clear in your previous posts, but it looks to me that you might have only provided information from the case.dayfile and the error files (cat *.error), but maybe not from the standard output. Are you still using the PBS script in your old post at http://www.mail-archive.com/wien%40zeus.theochem.tuwien.ac.at/msg11770.html ? In the script, I can see that the standard output is set to be written to a file called wien2k_output. When it runs fine on a single node, does it always use the same node (say z1-17) or does it run fine on other nodes (like z1-18)? I also tried changing the mpirun to mpiexec as suggested by Prof. Marks by setting: setenv WIEN_MPIRUN /usr/local/mvapich2-icc/bin/mpiexec -np _NP_ -f _HOSTS_ _EXEC_ in the parallel_option. In this case, the program does not run and also does not terminate (qstat on cluster just gives 00:00:00 for the time with a running status).. At least for now, stay with mpirun since it works on a single node. ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
Thanks for all the information and suggestions. I have tried to change -lmkl_blacs_intelmpi_lp64 to -lmkl_blacs_lp64 and recompile. However, I got the following error message in the screen output LAPW0 END [cli_14]: [cli_15]: [cli_6]: aborting job: Fatal error in PMPI_Comm_size: Invalid communicator, error stack: PMPI_Comm_size(110): MPI_Comm_size(comm=0x5b, size=0x7f190c) failed PMPI_Comm_size(69).: Invalid communicator aborting job: Fatal error in PMPI_Comm_size: Invalid communicator, error stack: PMPI_Comm_size(110): MPI_Comm_size(comm=0x5b, size=0x7f190c) failed PMPI_Comm_size(69).: Invalid communicator ... [z0-5:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 20. MPI process died? [z0-5:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z0-5:mpispawn_0][child_handler] MPI process (rank: 14, pid: 11260) exited with status 1 [z0-5:mpispawn_0][child_handler] MPI process (rank: 3, pid: 11249) exited with status 1 [z0-5:mpispawn_0][child_handler] MPI process (rank: 6, pid: 11252) exited with status 1 . Previously I compiled the program with -lmkl_blacs_intelmpi_lp64 and the mpi parallelization on a single node seems to be working. I notice that during the run, the *.error files have finite sizes, but I re-examine them after the job finished and there were no errors written inside (and the files have 0kb now). Does this indicates that the mpi is not running probably at all even on a single node? But I have checked the output result and it's in agreement with the non-mpi results..(for some simple cases) I also tried changing the mpirun to mpiexec as suggested by Prof. Marks by setting: setenv WIEN_MPIRUN /usr/local/mvapich2-icc/bin/mpiexec -np _NP_ -f _HOSTS_ _EXEC_ in the parallel_option. In this case, the program does not run and also does not terminate (qstat on cluster just gives 00:00:00 for the time with a running status).. Regards, Fermin ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
I have checked that case.vsp/vns are up-to-date. I guess lawp0_mpi runs properly. I compiled the source codes with ifort and please find the following for the linking options: current:FOPT:-FR -mp1 -w -prec_div -pc80 -pad -ip -DINTEL_VML -traceback current:FPOPT:-FR -mp1 -w -prec_div -pc80 -pad -ip -DINTEL_VML -Dmkl_scalapack -traceback current:FFTW_OPT:-DFFTW3 -I/usr/local/include current:FFTW_LIBS:-lfftw3_mpi -lfftw3 -L/usr/local/lib current:LDFLAGS:$(FOPT) -L/opt/intel/Compiler/11.1/046/mkl/lib/em64t -pthread current:DPARALLEL:'-DParallel' current:R_LIBS:-lmkl_lapack -lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core -openmp -lpthread -lguide current:RP_LIBS:-lmkl_scalapack_lp64 -lmkl_solver_lp64 -lmkl_blacs_intelmpi_lp64 $(R_LIBS) current:MPIRUN:/usr/local/mvapich2-icc/bin/mpirun -np _NP_ -hostfile _HOSTS_ _EXEC_ current:MKL_TARGET_ARCH:intel64 Is it ok to use -lmkl_blacs_intelmpi_lp64? Thanks a lot for all the suggestions. Regards, Fermin -Original Message- From: wien-boun...@zeus.theochem.tuwien.ac.at [mailto: wien-boun...@zeus.theochem.tuwien.ac.at] On Behalf Of Peter Blaha To: A Mailing list for WIEN2k users Subject: Re: [Wien] Error in mpi+k point parallelization across multiple nodes It seems as if lapw0_mpi runs properly ?? Please check if you have NEW (check date with ls -als)!! valid case.vsp/vns files, which can be used in eg. a sequential lapw1 step. This suggests that mpi and fftw are ok. The problems seem to start in lapw1_mpi, and this program requires in addition to mpi also scalapack. I guess you compile with ifort and link with the mkl ?? There is one crucial blacs library, which must be adapted to your mpi, since they are specific to a particular mpi (intelmpi, openmpi, ...): Which blacks-library do you link ? -lmkl_blacs_lp64 or another one ?? Check out the doku for the mkl. Am 04.05.2015 um 05:18 schrieb lung Fermin: I have tried to set MPI_REMOTE=0 and used 32 cores (on 2 nodes) for distributing the mpi job. However, the problem still persist... but the error message looks different this time: $ cat *.error Error in LAPW2 ** testerror: Error in Parallel LAPW2 and the output on screen: Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-1 8 z1-18 z1-18 number of processors: 32 LAPW0 END [16] Failed to dealloc pd (Device or resource busy) [0] Failed to dealloc pd (Device or resource busy) [17] Failed to dealloc pd (Device or resource busy) [2] Failed to dealloc pd (Device or resource busy) [18] Failed to dealloc pd (Device or resource busy) [1] Failed to dealloc pd (Device or resource busy) LAPW1 END LAPW2 - FERMI; weighs written [z1-17:mpispawn_0][child_handler] MPI process (rank: 0, pid: 28291) terminated with signal 9 - abort job [z1-17:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 9. MPI process died? [z1-17:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-17:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-17 aborted: Error while reading a PMI socket (4) [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][handle_mt_peer] Error while reading PMI socket. MPI process died? cp: cannot stat `.in.tmp': No such file or directory stop error -- -- ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
No !!! (Can use it only if you are using intelmpi). I'm not sure (and it may even depend on the compiler version) which mpi-versions are supported by intel. But maybe try the simplest version -lmkl_blacs_lp64 Am 04.05.2015 um 08:03 schrieb lung Fermin: Is it ok to use -lmkl_blacs_intelmpi_lp64? -- - Peter Blaha Inst. Materials Chemistry, TU Vienna Getreidemarkt 9, A-1060 Vienna, Austria Tel: +43-1-5880115671 Fax: +43-1-5880115698 email: pbl...@theochem.tuwien.ac.at - ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
On page 131 in the User's Guide for Intel mkl 11.1 for Linux [ https://software.intel.com/en-us/mkl_11.1_ug_lin_pdf ], it has: libmkl_blacs_intelmpi_lp64.so = LP64 version of BLACS routines for Intel MPI and MPICH2 So -lmkl_blacs_intelmpi_lp64 might also work with MPICH2. From the compile settings, it looks like compiler version 11.1 build 46 is being used, which uses version 10.2 Update 2 of mkl [ https://software.intel.com/en-us/articles/which-version-of-ipp--mkl--tbb-is-installed-with-intel-compiler-professional-edition ]. I could not find the mkl system requirements page for 10.2, but for 10.3, it has that 10.3 was validated with MPICH2 version 1.3.2p1 [ https://software.intel.com/en-us/articles/intel-mkl-103-system-requirements ]. The MVAPICH2-2.0a that was mentioned as being used is based on MPICH-3.0.4 [ http://mvapich.cse.ohio-state.edu/static/media/mvapich/MV2_CHANGELOG-2.1.txt ]. It looks like Intel did not validate MPICH3 until about mkl version 11.2 with MPICH3 version 3.1 [ https://software.intel.com/en-us/articles/intel-mkl-112-system-requirements ]. Fermin, it looks like you provided information from the WIEN2k dayfile and error files. However, are there any error messages in the standard output? For example, standard output should be what you get in a terminal after you execute runsp_lapw -p. However, since clusters typically require that calculations be submitted using a job submission system, the standard output is usually written instead to a user specified file. On some systems, for example, the calculation isn't executed with runsp_lapw -p but with qsub -j oe -o output.log myscript.pbs [ http://arc.it.wsu.edu/UserGuide/Using_Qsub.aspx ], where the script file called myscript.pbs would contain a line to execute runsp_lapw -p and the standard output would be written to the file called output.log. On 5/4/2015 12:08 AM, Peter Blaha wrote: No !!! (Can use it only if you are using intelmpi). I'm not sure (and it may even depend on the compiler version) which mpi-versions are supported by intel. But maybe try the simplest version -lmkl_blacs_lp64 Am 04.05.2015 um 08:03 schrieb lung Fermin: Is it ok to use -lmkl_blacs_intelmpi_lp64? ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
To reiterate what everyone else said, you should change your blacs, the intelmpi version only works if you are using impi (I am 98% certain). Normally this leads to a wierd but understandable error when lapw0/lapw1 initiate the mpi routines, not sure why this did not show up in your case. On Mon, May 4, 2015 at 12:56 PM, Gavin Abo gs...@crimson.ua.edu wrote: On page 131 in the User's Guide for Intel mkl 11.1 for Linux [ https://software.intel.com/en-us/mkl_11.1_ug_lin_pdf ], it has: libmkl_blacs_intelmpi_lp64.so = LP64 version of BLACS routines for Intel MPI and MPICH2 So -lmkl_blacs_intelmpi_lp64 might also work with MPICH2. From the compile settings, it looks like compiler version 11.1 build 46 is being used, which uses version 10.2 Update 2 of mkl [ https://software.intel.com/en-us/articles/which-version-of-ipp--mkl--tbb-is-installed-with-intel-compiler-professional-edition ]. I could not find the mkl system requirements page for 10.2, but for 10.3, it has that 10.3 was validated with MPICH2 version 1.3.2p1 [ https://software.intel.com/en-us/articles/intel-mkl-103-system-requirements ]. The MVAPICH2-2.0a that was mentioned as being used is based on MPICH-3.0.4 [ http://mvapich.cse.ohio-state.edu/static/media/mvapich/MV2_CHANGELOG-2.1.txt ]. It looks like Intel did not validate MPICH3 until about mkl version 11.2 with MPICH3 version 3.1 [ https://software.intel.com/en-us/articles/intel-mkl-112-system-requirements ]. Fermin, it looks like you provided information from the WIEN2k dayfile and error files. However, are there any error messages in the standard output? For example, standard output should be what you get in a terminal after you execute runsp_lapw -p. However, since clusters typically require that calculations be submitted using a job submission system, the standard output is usually written instead to a user specified file. On some systems, for example, the calculation isn't executed with runsp_lapw -p but with qsub -j oe -o output.log myscript.pbs [ http://arc.it.wsu.edu/UserGuide/Using_Qsub.aspx ], where the script file called myscript.pbs would contain a line to execute runsp_lapw -p and the standard output would be written to the file called output.log. On 5/4/2015 12:08 AM, Peter Blaha wrote: No !!! (Can use it only if you are using intelmpi). I'm not sure (and it may even depend on the compiler version) which mpi-versions are supported by intel. But maybe try the simplest version -lmkl_blacs_lp64 Am 04.05.2015 um 08:03 schrieb lung Fermin: Is it ok to use -lmkl_blacs_intelmpi_lp64? ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html -- Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu Corrosion in 4D: MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
I have tried to set MPI_REMOTE=0 and used 32 cores (on 2 nodes) for distributing the mpi job. However, the problem still persist... but the error message looks different this time: $ cat *.error Error in LAPW2 ** testerror: Error in Parallel LAPW2 and the output on screen: Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-1 8 z1-18 z1-18 number of processors: 32 LAPW0 END [16] Failed to dealloc pd (Device or resource busy) [0] Failed to dealloc pd (Device or resource busy) [17] Failed to dealloc pd (Device or resource busy) [2] Failed to dealloc pd (Device or resource busy) [18] Failed to dealloc pd (Device or resource busy) [1] Failed to dealloc pd (Device or resource busy) LAPW1 END LAPW2 - FERMI; weighs written [z1-17:mpispawn_0][child_handler] MPI process (rank: 0, pid: 28291) terminated with signal 9 - abort job [z1-17:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 9. MPI process died? [z1-17:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-17:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-17 aborted: Error while reading a PMI socket (4) [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][handle_mt_peer] Error while reading PMI socket. MPI process died? cp: cannot stat `.in.tmp': No such file or directory stop error Try setting setenv MPI_REMOTE 0 in parallel options. Am 29.04.2015 um 09:44 schrieb lung Fermin: Thanks for your comment, Prof. Marks. Each node on the cluster has 32GB memory and each core (16 in total) on the node is limited to 2GB of memory usage. For the current system, I used RKMAX=6, and the smallest RMT=2.25. I have tested the calculation with single k point and mpi on 16 cores within a node. The matrix size from $ cat *.nmat_only is 29138 Does this mean that the number of matrix elements is 29138 or (29138)^2? In general, how shall I estimate the memory required for a calculation? I have also checked the memory usage with top on the node. Each core has used up ~5% of the memory and this adds up to ~5*16% on the node. Perhaps the problem is really caused by the overflow of memory.. I am now queuing on the cluster to test for the case of mpi over 32 cores (2 nodes). Thanks. Regards, Fermin -- -- As an addendum, the calculation may be too big for a single node. How much memory does the node have, what is the RKMAX, the smallest RMT unit cell size? Maybe use in your machines file 1:z1-2:16 z1-13:16 lapw0: z1-2:16 z1-13:16 granularity:1 extrafine:1 Check the size using x law1 -c -p -nmat_only cat *.nmat ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu http://www.numis.northwestern.edu MURI4D.numis.northwestern.edu http://MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:45 PM, Laurence Marks l-ma...@northwestern.edu mailto:l-ma...@northwestern.edu l-ma...@northwestern.edu wrote: Unfortunately it is hard to know what is going on. A google search on Error while reading PMI socket. indicates that the message you have means it did not work, and is not specific. Some suggestions: a) Try mpiexec (slightly different arguments). You just edit parallel_options. https://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager b) Try an older version of mvapich2 if it is on the system. c) Do you have to launch mpdboot for your system https://wiki.calculquebec.ca/w/MVAPICH2/en? d) Talk to a sys_admin, particularly the one who setup mvapich e) Do cat *.error, maybe something else went wrong or it is not mpi's fault but a user error. ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu http://www.numis.northwestern.edu MURI4D.numis.northwestern.edu http://MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:17 PM, lung Fermin ferminl...@gmail.com mailto:ferminl...@gmail.com ferminl...@gmail.com wrote:
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
It seems as if lapw0_mpi runs properly ?? Please check if you have NEW (check date with ls -als)!! valid case.vsp/vns files, which can be used in eg. a sequential lapw1 step. This suggests that mpi and fftw are ok. The problems seem to start in lapw1_mpi, and this program requires in addition to mpi also scalapack. I guess you compile with ifort and link with the mkl ?? There is one crucial blacs library, which must be adapted to your mpi, since they are specific to a particular mpi (intelmpi, openmpi, ...): Which blacks-library do you link ? -lmkl_blacs_lp64 or another one ?? Check out the doku for the mkl. Am 04.05.2015 um 05:18 schrieb lung Fermin: I have tried to set MPI_REMOTE=0 and used 32 cores (on 2 nodes) for distributing the mpi job. However, the problem still persist... but the error message looks different this time: $ cat *.error Error in LAPW2 ** testerror: Error in Parallel LAPW2 and the output on screen: Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-17 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-18 z1-1 8 z1-18 z1-18 number of processors: 32 LAPW0 END [16] Failed to dealloc pd (Device or resource busy) [0] Failed to dealloc pd (Device or resource busy) [17] Failed to dealloc pd (Device or resource busy) [2] Failed to dealloc pd (Device or resource busy) [18] Failed to dealloc pd (Device or resource busy) [1] Failed to dealloc pd (Device or resource busy) LAPW1 END LAPW2 - FERMI; weighs written [z1-17:mpispawn_0][child_handler] MPI process (rank: 0, pid: 28291) terminated with signal 9 - abort job [z1-17:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 9. MPI process died? [z1-17:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-17:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-17 aborted: Error while reading a PMI socket (4) [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 21. MPI process died? [z1-18:mpispawn_1][handle_mt_peer] Error while reading PMI socket. MPI process died? cp: cannot stat `.in.tmp': No such file or directory stop error Try setting setenv MPI_REMOTE 0 in parallel options. Am 29.04.2015 um 09:44 schrieb lung Fermin: Thanks for your comment, Prof. Marks. Each node on the cluster has 32GB memory and each core (16 in total) on the node is limited to 2GB of memory usage. For the current system, I used RKMAX=6, and the smallest RMT=2.25. I have tested the calculation with single k point and mpi on 16 cores within a node. The matrix size from $ cat *.nmat_only is 29138 Does this mean that the number of matrix elements is 29138 or (29138)^2? In general, how shall I estimate the memory required for a calculation? I have also checked the memory usage with top on the node. Each core has used up ~5% of the memory and this adds up to ~5*16% on the node. Perhaps the problem is really caused by the overflow of memory.. I am now queuing on the cluster to test for the case of mpi over 32 cores (2 nodes). Thanks. Regards, Fermin -- -- As an addendum, the calculation may be too big for a single node. How much memory does the node have, what is the RKMAX, the smallest RMT unit cell size? Maybe use in your machines file 1:z1-2:16 z1-13:16 lapw0: z1-2:16 z1-13:16 granularity:1 extrafine:1 Check the size using x law1 -c -p -nmat_only cat *.nmat ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern Universitywww.numis.northwestern.edu http://www.numis.northwestern.edu http://www.numis.northwestern.edu MURI4D.numis.northwestern.edu http://MURI4D.numis.northwestern.edu http://MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:45 PM, Laurence Marks l-ma...@northwestern.edu mailto:l-ma...@northwestern.edu mailto:l-ma...@northwestern.edu wrote: Unfortunately it is hard to know what is going on. A google search on Error while reading PMI socket. indicates that the message you have means it did not work, and is not specific. Some suggestions: a) Try mpiexec (slightly different arguments). You just edit parallel_options.
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
Thanks for your comment, Prof. Marks. Each node on the cluster has 32GB memory and each core (16 in total) on the node is limited to 2GB of memory usage. For the current system, I used RKMAX=6, and the smallest RMT=2.25. I have tested the calculation with single k point and mpi on 16 cores within a node. The matrix size from $ cat *.nmat_only is 29138 Does this mean that the number of matrix elements is 29138 or (29138)^2? In general, how shall I estimate the memory required for a calculation? I have also checked the memory usage with top on the node. Each core has used up ~5% of the memory and this adds up to ~5*16% on the node. Perhaps the problem is really caused by the overflow of memory.. I am now queuing on the cluster to test for the case of mpi over 32 cores (2 nodes). Thanks. Regards, Fermin As an addendum, the calculation may be too big for a single node. How much memory does the node have, what is the RKMAX, the smallest RMT unit cell size? Maybe use in your machines file 1:z1-2:16 z1-13:16 lapw0: z1-2:16 z1-13:16 granularity:1 extrafine:1 Check the size using x law1 -c -p -nmat_only cat *.nmat ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:45 PM, Laurence Marks l-ma...@northwestern.edu wrote: Unfortunately it is hard to know what is going on. A google search on Error while reading PMI socket. indicates that the message you have means it did not work, and is not specific. Some suggestions: a) Try mpiexec (slightly different arguments). You just edit parallel_options. https://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager b) Try an older version of mvapich2 if it is on the system. c) Do you have to launch mpdboot for your system https://wiki.calculquebec.ca/w/MVAPICH2/en? d) Talk to a sys_admin, particularly the one who setup mvapich e) Do cat *.error, maybe something else went wrong or it is not mpi's fault but a user error. ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:17 PM, lung Fermin ferminl...@gmail.com wrote: Thanks for Prof. Marks' comment. 1. In the previous email, I have missed to copy the line setenv WIEN_MPIRUN /usr/local/mvapich2-icc/bin/mpirun -np _NP_ -hostfile _HOSTS_ _EXEC_ It was in the parallel_option. Sorry about that. 2. I have checked that the running program was lapw1c_mpi. Besides, when the mpi calculation was done on a single node for some other system, the results are consistent with the literature. So I believe that the mpi code has been setup and compiled properly. Would there be something wrong with my option in siteconfig..? Do I have to set some command to bind the job? Any other possible cause of the error? Any suggestions or comments would be appreciated. Thanks. Regards, Fermin You appear to be missing the line setenv WIEN_MPIRUN=... This is setup when you run siteconfig, and provides the information on how mpi is run on your system. N.B., did you setup and compile the mpi code? ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 4:22 AM, lung Fermin ferminl...@gmail.com wrote: Dear Wien2k community, I am trying to perform calculation on a system of ~100 in-equivalent atoms using mpi+k point parallelization on a cluster. Everything goes fine when the program was run on a single node. However, if I perform the calculation across different nodes, the follow error occurs. How to solve this problem? I am a newbie to mpi programming, any help would be appreciated. Thanks. The error message (MVAPICH2 2.0a): --- Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1 -13 z1-13 z1-13 z1-13 z1-13 z1-13 number of processors: 32 LAPW0 END
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
As an addendum, the calculation may be too big for a single node. How much memory does the node have, what is the RKMAX, the smallest RMT unit cell size? Maybe use in your machines file 1:z1-2:16 z1-13:16 lapw0: z1-2:16 z1-13:16 granularity:1 extrafine:1 Check the size using x law1 -c -p -nmat_only cat *.nmat ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:45 PM, Laurence Marks l-ma...@northwestern.edu wrote: Unfortunately it is hard to know what is going on. A google search on Error while reading PMI socket. indicates that the message you have means it did not work, and is not specific. Some suggestions: a) Try mpiexec (slightly different arguments). You just edit parallel_options. https://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager b) Try an older version of mvapich2 if it is on the system. c) Do you have to launch mpdboot for your system https://wiki.calculquebec.ca/w/MVAPICH2/en? d) Talk to a sys_admin, particularly the one who setup mvapich e) Do cat *.error, maybe something else went wrong or it is not mpi's fault but a user error. ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:17 PM, lung Fermin ferminl...@gmail.com wrote: Thanks for Prof. Marks' comment. 1. In the previous email, I have missed to copy the line setenv WIEN_MPIRUN /usr/local/mvapich2-icc/bin/mpirun -np _NP_ -hostfile _HOSTS_ _EXEC_ It was in the parallel_option. Sorry about that. 2. I have checked that the running program was lapw1c_mpi. Besides, when the mpi calculation was done on a single node for some other system, the results are consistent with the literature. So I believe that the mpi code has been setup and compiled properly. Would there be something wrong with my option in siteconfig..? Do I have to set some command to bind the job? Any other possible cause of the error? Any suggestions or comments would be appreciated. Thanks. Regards, Fermin You appear to be missing the line setenv WIEN_MPIRUN=... This is setup when you run siteconfig, and provides the information on how mpi is run on your system. N.B., did you setup and compile the mpi code? ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 4:22 AM, lung Fermin ferminl...@gmail.com wrote: Dear Wien2k community, I am trying to perform calculation on a system of ~100 in-equivalent atoms using mpi+k point parallelization on a cluster. Everything goes fine when the program was run on a single node. However, if I perform the calculation across different nodes, the follow error occurs. How to solve this problem? I am a newbie to mpi programming, any help would be appreciated. Thanks. The error message (MVAPICH2 2.0a): --- Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1 -13 z1-13 z1-13 z1-13 z1-13 z1-13 number of processors: 32 LAPW0 END [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-13 aborted: Error while reading a PMI socket (4) [z1-13:mpispawn_0][child_handler] MPI process (rank: 11, pid: 8546) terminated with signal 9 - abort job [z1-13:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 8. MPI process died? [z1-13:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 12. MPI process died? [z1-2:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][child_handler] MPI process (rank: 0, pid: 35454) terminated with signal 9 - abort job [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-2 aborted: MPI process error (1) [cli_15]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 15 stop
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
Try setting setenv MPI_REMOTE 0 in parallel options. Am 29.04.2015 um 09:44 schrieb lung Fermin: Thanks for your comment, Prof. Marks. Each node on the cluster has 32GB memory and each core (16 in total) on the node is limited to 2GB of memory usage. For the current system, I used RKMAX=6, and the smallest RMT=2.25. I have tested the calculation with single k point and mpi on 16 cores within a node. The matrix size from $ cat *.nmat_only is 29138 Does this mean that the number of matrix elements is 29138 or (29138)^2? In general, how shall I estimate the memory required for a calculation? I have also checked the memory usage with top on the node. Each core has used up ~5% of the memory and this adds up to ~5*16% on the node. Perhaps the problem is really caused by the overflow of memory.. I am now queuing on the cluster to test for the case of mpi over 32 cores (2 nodes). Thanks. Regards, Fermin As an addendum, the calculation may be too big for a single node. How much memory does the node have, what is the RKMAX, the smallest RMT unit cell size? Maybe use in your machines file 1:z1-2:16 z1-13:16 lapw0: z1-2:16 z1-13:16 granularity:1 extrafine:1 Check the size using x law1 -c -p -nmat_only cat *.nmat ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu http://www.numis.northwestern.edu MURI4D.numis.northwestern.edu http://MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:45 PM, Laurence Marks l-ma...@northwestern.edu mailto:l-ma...@northwestern.edu wrote: Unfortunately it is hard to know what is going on. A google search on Error while reading PMI socket. indicates that the message you have means it did not work, and is not specific. Some suggestions: a) Try mpiexec (slightly different arguments). You just edit parallel_options. https://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager b) Try an older version of mvapich2 if it is on the system. c) Do you have to launch mpdboot for your system https://wiki.calculquebec.ca/w/MVAPICH2/en? d) Talk to a sys_admin, particularly the one who setup mvapich e) Do cat *.error, maybe something else went wrong or it is not mpi's fault but a user error. ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu http://www.numis.northwestern.edu MURI4D.numis.northwestern.edu http://MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:17 PM, lung Fermin ferminl...@gmail.com mailto:ferminl...@gmail.com wrote: Thanks for Prof. Marks' comment. 1. In the previous email, I have missed to copy the line setenv WIEN_MPIRUN /usr/local/mvapich2-icc/bin/mpirun -np _NP_ -hostfile _HOSTS_ _EXEC_ It was in the parallel_option. Sorry about that. 2. I have checked that the running program was lapw1c_mpi. Besides, when the mpi calculation was done on a single node for some other system, the results are consistent with the literature. So I believe that the mpi code has been setup and compiled properly. Would there be something wrong with my option in siteconfig..? Do I have to set some command to bind the job? Any other possible cause of the error? Any suggestions or comments would be appreciated. Thanks. Regards, Fermin You appear to be missing the line setenv WIEN_MPIRUN=... This is setup when you run siteconfig, and provides the information on how mpi is run on your system. N.B., did you setup and compile the mpi code? ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu http://www.numis.northwestern.edu MURI4D.numis.northwestern.edu http://MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 4:22 AM, lung Fermin ferminl...@gmail.com mailto:ferminl...@gmail.com wrote: Dear Wien2k community, I am trying to perform calculation on a system of ~100 in-equivalent atoms using mpi+k point parallelization on a cluster. Everything goes fine when the program was run on a single node. However, if I perform the calculation across different nodes, the follow error occurs. How to solve this problem? I am a newbie to mpi programming, any help would be appreciated. Thanks. The error message (MVAPICH2 2.0a):
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
You appear to be missing the line setenv WIEN_MPIRUN=... This is setup when you run siteconfig, and provides the information on how mpi is run on your system. N.B., did you setup and compile the mpi code? ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 4:22 AM, lung Fermin ferminl...@gmail.com wrote: Dear Wien2k community, I am trying to perform calculation on a system of ~100 in-equivalent atoms using mpi+k point parallelization on a cluster. Everything goes fine when the program was run on a single node. However, if I perform the calculation across different nodes, the follow error occurs. How to solve this problem? I am a newbie to mpi programming, any help would be appreciated. Thanks. The error message (MVAPICH2 2.0a): --- Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1 -13 z1-13 z1-13 z1-13 z1-13 z1-13 number of processors: 32 LAPW0 END [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-13 aborted: Error while reading a PMI socket (4) [z1-13:mpispawn_0][child_handler] MPI process (rank: 11, pid: 8546) terminated with signal 9 - abort job [z1-13:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 8. MPI process died? [z1-13:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 12. MPI process died? [z1-2:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][child_handler] MPI process (rank: 0, pid: 35454) terminated with signal 9 - abort job [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-2 aborted: MPI process error (1) [cli_15]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 15 stop error -- The .machines file: # 1:z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 1:z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 granularity:1 extrafine:1 The parallel_options: setenv TASKSET no setenv USE_REMOTE 0 setenv MPI_REMOTE 1 setenv WIEN_GRANULARITY 1 Thanks. Regards, Fermin ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
Thanks for Prof. Marks' comment. 1. In the previous email, I have missed to copy the line setenv WIEN_MPIRUN /usr/local/mvapich2-icc/bin/mpirun -np _NP_ -hostfile _HOSTS_ _EXEC_ It was in the parallel_option. Sorry about that. 2. I have checked that the running program was lapw1c_mpi. Besides, when the mpi calculation was done on a single node for some other system, the results are consistent with the literature. So I believe that the mpi code has been setup and compiled properly. Would there be something wrong with my option in siteconfig..? Do I have to set some command to bind the job? Any other possible cause of the error? Any suggestions or comments would be appreciated. Thanks. Regards, Fermin You appear to be missing the line setenv WIEN_MPIRUN=... This is setup when you run siteconfig, and provides the information on how mpi is run on your system. N.B., did you setup and compile the mpi code? ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 4:22 AM, lung Fermin ferminl...@gmail.com wrote: Dear Wien2k community, I am trying to perform calculation on a system of ~100 in-equivalent atoms using mpi+k point parallelization on a cluster. Everything goes fine when the program was run on a single node. However, if I perform the calculation across different nodes, the follow error occurs. How to solve this problem? I am a newbie to mpi programming, any help would be appreciated. Thanks. The error message (MVAPICH2 2.0a): --- Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1 -13 z1-13 z1-13 z1-13 z1-13 z1-13 number of processors: 32 LAPW0 END [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-13 aborted: Error while reading a PMI socket (4) [z1-13:mpispawn_0][child_handler] MPI process (rank: 11, pid: 8546) terminated with signal 9 - abort job [z1-13:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 8. MPI process died? [z1-13:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 12. MPI process died? [z1-2:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][child_handler] MPI process (rank: 0, pid: 35454) terminated with signal 9 - abort job [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-2 aborted: MPI process error (1) [cli_15]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 15 stop error -- The .machines file: # 1:z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 1:z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 granularity:1 extrafine:1 The parallel_options: setenv TASKSET no setenv USE_REMOTE 0 setenv MPI_REMOTE 1 setenv WIEN_GRANULARITY 1 Thanks. Regards, Fermin ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
Re: [Wien] Error in mpi+k point parallelization across multiple nodes
Unfortunately it is hard to know what is going on. A google search on Error while reading PMI socket. indicates that the message you have means it did not work, and is not specific. Some suggestions: a) Try mpiexec (slightly different arguments). You just edit parallel_options. https://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager b) Try an older version of mvapich2 if it is on the system. c) Do you have to launch mpdboot for your system https://wiki.calculquebec.ca/w/MVAPICH2/en? d) Talk to a sys_admin, particularly the one who setup mvapich e) Do cat *.error, maybe something else went wrong or it is not mpi's fault but a user error. ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 10:17 PM, lung Fermin ferminl...@gmail.com wrote: Thanks for Prof. Marks' comment. 1. In the previous email, I have missed to copy the line setenv WIEN_MPIRUN /usr/local/mvapich2-icc/bin/mpirun -np _NP_ -hostfile _HOSTS_ _EXEC_ It was in the parallel_option. Sorry about that. 2. I have checked that the running program was lapw1c_mpi. Besides, when the mpi calculation was done on a single node for some other system, the results are consistent with the literature. So I believe that the mpi code has been setup and compiled properly. Would there be something wrong with my option in siteconfig..? Do I have to set some command to bind the job? Any other possible cause of the error? Any suggestions or comments would be appreciated. Thanks. Regards, Fermin You appear to be missing the line setenv WIEN_MPIRUN=... This is setup when you run siteconfig, and provides the information on how mpi is run on your system. N.B., did you setup and compile the mpi code? ___ Professor Laurence Marks Department of Materials Science and Engineering Northwestern University www.numis.northwestern.edu MURI4D.numis.northwestern.edu Co-Editor, Acta Cryst A Research is to see what everybody else has seen, and to think what nobody else has thought Albert Szent-Gyorgi On Apr 28, 2015 4:22 AM, lung Fermin ferminl...@gmail.com wrote: Dear Wien2k community, I am trying to perform calculation on a system of ~100 in-equivalent atoms using mpi+k point parallelization on a cluster. Everything goes fine when the program was run on a single node. However, if I perform the calculation across different nodes, the follow error occurs. How to solve this problem? I am a newbie to mpi programming, any help would be appreciated. Thanks. The error message (MVAPICH2 2.0a): --- Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1 -13 z1-13 z1-13 z1-13 z1-13 z1-13 number of processors: 32 LAPW0 END [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-13 aborted: Error while reading a PMI socket (4) [z1-13:mpispawn_0][child_handler] MPI process (rank: 11, pid: 8546) terminated with signal 9 - abort job [z1-13:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 8. MPI process died? [z1-13:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 12. MPI process died? [z1-2:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][child_handler] MPI process (rank: 0, pid: 35454) terminated with signal 9 - abort job [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-2 aborted: MPI process error (1) [cli_15]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 15 stop error -- The .machines file: # 1:z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 1:z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 granularity:1 extrafine:1 The parallel_options: setenv TASKSET no setenv USE_REMOTE 0 setenv MPI_REMOTE 1 setenv WIEN_GRANULARITY 1 Thanks. Regards, Fermin
[Wien] Error in mpi+k point parallelization across multiple nodes
Dear Wien2k community, I am trying to perform calculation on a system of ~100 in-equivalent atoms using mpi+k point parallelization on a cluster. Everything goes fine when the program was run on a single node. However, if I perform the calculation across different nodes, the follow error occurs. How to solve this problem? I am a newbie to mpi programming, any help would be appreciated. Thanks. The error message (MVAPICH2 2.0a): --- Warning: no access to tty (Bad file descriptor). Thus no job control in this shell. z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1 -13 z1-13 z1-13 z1-13 z1-13 z1-13 number of processors: 32 LAPW0 END [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-13 aborted: Error while reading a PMI socket (4) [z1-13:mpispawn_0][child_handler] MPI process (rank: 11, pid: 8546) terminated with signal 9 - abort job [z1-13:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 8. MPI process died? [z1-13:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][readline] Unexpected End-Of-File on file descriptor 12. MPI process died? [z1-2:mpispawn_0][mtpmi_processops] Error while reading PMI socket. MPI process died? [z1-2:mpispawn_0][child_handler] MPI process (rank: 0, pid: 35454) terminated with signal 9 - abort job [z1-2:mpirun_rsh][process_mpispawn_connection] mpispawn_0 from node z1-2 aborted: MPI process error (1) [cli_15]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 15 stop error -- The .machines file: # 1:z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 z1-2 1:z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 z1-13 granularity:1 extrafine:1 The parallel_options: setenv TASKSET no setenv USE_REMOTE 0 setenv MPI_REMOTE 1 setenv WIEN_GRANULARITY 1 Thanks. Regards, Fermin ___ Wien mailing list Wien@zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien SEARCH the MAILING-LIST at: http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html