April 2014 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] MPICH (v3.1) Performing slow on 10G Ethernet
by Muhammad Ansar Javed 17 Apr '14

17 Apr '14

I am running bandwidth test from OSU-Micro-Benchmarks and a simple Ping-Pong program implemented using MPI Send and Recv methods. I have also tested bandwidth with NetPipe MPI implementation as well but no improvements in results. Interrupt coalescing was ON by default and disabling it makes latency consistent at 37 us through out messages of 1 byte to 2 MB data size. I have done following vendor recommended optimizations for network buffer. Latency is not increasing with the message size may be because of these settings. net.core.rmem_max = 16777216 net.core.wmem_max = 16777216 net.ipv4.tcp_rmem = 4096 87380 16777216 net.ipv4.tcp_wmem = 4096 65536 16777216 net.core.netdev_max_backlog = 250000 I have not run streaming benchmark. I found a Streaming MPI benchmark ( http://www.cs.virginia.edu/stream/) but its C implementation is not available. Can you please suggest me any Streaming MPI-C benchmark? Latency test with netperf TCP_RR gives 37 us for message size 1 Bytes. Moreover, according to Netperf docs, _RR test support only 1 bytes message size. Please correct me if I am wrong. Thanks, On Wed, Apr 16, 2014 at 5:36 PM, Atchley, Scott <atchleyes(a)ornl.gov> wrote: > On Apr 16, 2014, at 8:35 AM, Scott Atchley <atchleyes(a)ornl.gov> wrote: > > > Keep in mind that iperf is a streaming test. What MPI application are > you running? IMB pingpong? > > > > You should review you NIC vendor's tuning FAQ. Is interrupt coalescing > turned on? Normally it is, but it increases small message latency. > > > > As for bandwidth, have you tried a streaming MPI benchmark? > > Or run Netperf's TCP_RR for each message size to see how it compares to > pingpong. > > > > > Scott > > > > On Apr 15, 2014, at 4:29 PM, Muhammad Ansar Javed < > muhammad.ansar(a)seecs.edu.pk> wrote: > > > >> Hi Antonio, > >> Thanks for response. > >> Here is output of iperf bandwidth test. > >> . > >> Client connecting to 10g2, TCP port 5001 > >> TCP window size: 1.41 MByte (default) > >> ------------------------------------------------------------ > >> [ 5] local 192.168.1.33 port 34543 connected with 192.168.1.34 port > 5001 > >> [ 4] local 192.168.1.33 port 5001 connected with 192.168.1.34 port > 48531 > >> [ ID] Interval Transfer Bandwidth > >> [ 5] 0.0-10.0 sec 11.2 GBytes 9.61 Gbits/sec > >> [ 4] 0.0-10.0 sec 11.2 GBytes 9.59 Gbits/sec > >> > >> > >> > >> On Tue, Apr 15, 2014 at 7:17 PM, "Antonio J. Peña" <apenya(a)mcs.anl.gov> > wrote: > >> > >> Hi Ansar, > >> > >> The bandwidth numbers seem a little low. I'd expect something around > 7,000 Mbps. To discard issues with MPI, you can try generic benchmarking > tools such as those described here> > >> > >> http://linuxaria.com/article/tool-command-line-bandwidth-linux?lang=en > >> > >> If you get similar numbers, you'll have to check your setup. Otherwise, > let us know and we'll try to see if there's any performance issue with > MPICH. > >> > >> Best, > >> Antonio > >> > >> > >> > >> On 04/15/2014 06:35 AM, Muhammad Ansar Javed wrote: > >>> Hi, > >>> I am running benchmarks for MPICH v3.1 performance evaluation on 10G > Ethernet connection between two hosts. The performance results are less > than expected. Here is complete set of numbers for Latency and Bandwidth > tests. > >>> > >>> mpj@host3:~/code/benchmarks$ mpiexec -n 2 -f machines ./latency.out > >>> Latency: MPICH-C > >>> Size ______ Time (us) > >>> 1 _________ 37.91 > >>> 2 _________ 37.76 > >>> 4 _________ 37.75 > >>> 8 _________ 39.02 > >>> 16 ________ 52.71 > >>> 32 ________ 39.08 > >>> 64 ________ 37.75 > >>> 128 _______ 37.77 > >>> 256 _______ 57.47 > >>> 512 _______ 37.86 > >>> 1024 ______ 37.76 > >>> 2048 ______ 37.88 > >>> > >>> mpj@host3:~/code/benchmarks$ mpiexec -n 2 -f machines ./bandwidth.out > >>> Bandwidth: MPICH-C > >>> Size(Bytes) Bandwidth (Mbps) > >>> 2048 ______ 412.32 > >>> 4096 ______ 820.06 > >>> 8192 ______ 827.77 > >>> 16384 _____ 1644.36 > >>> 32768 _____ 2207.52 > >>> 65536 _____ 4368.76 > >>> 131072 ____ 2942.93 > >>> 262144 ____ 4281.17 > >>> 524288 ____ 4773.78 > >>> 1048576 ___ 5310.85 > >>> 2097152 ___ 5382.94 > >>> 4194304 ___ 5518.97 > >>> 8388608 ___ 5508.87 > >>> 16777216 __ 5498.93 > >>> > >>> My environments consists of two hosts having point-to-point > (switch-less) 10Gbps Ethernet connection. Environment (OS, user, directory > structure etc) on both hosts is exactly same. There is no NAS or shared > file system between hosts. I have attached output of script > >>> mpiexec -all. > >>> > >>> Are these numbers okay? If not then please suggest performance > improvement methodology... > >>> > >>> Thanks > >>> > >>> -- > >>> Regards > >>> > >>> Ansar Javed > >>> HPC Lab > >>> SEECS NUST > >>> Contact: +92 334 438 9394 > >>> Email: muhammad.ansar(a)seecs.edu.pk > >>> > >>> > >>> _______________________________________________ > >>> discuss mailing list > >>> discuss(a)mpich.org > >>> > >>> To manage subscription options or unsubscribe: > >>> > >>> https://lists.mpich.org/mailman/listinfo/discuss > >> > >> > >> -- > >> Antonio J. Peña > >> Postdoctoral Appointee > >> Mathematics and Computer Science Division > >> Argonne National Laboratory > >> 9700 South Cass Avenue, Bldg. 240, Of. 3148 > >> Argonne, IL 60439-4847 > >> > >> apenya(a)mcs.anl.gov > >> www.mcs.anl.gov/~apenya > >> > >> _______________________________________________ > >> discuss mailing list discuss(a)mpich.org > >> To manage subscription options or unsubscribe: > >> https://lists.mpich.org/mailman/listinfo/discuss > >> > >> > >> > >> -- > >> Regards > >> > >> Ansar Javed > >> HPC Lab > >> SEECS NUST > >> Contact: +92 334 438 9394 > >> Email: muhammad.ansar(a)seecs.edu.pk > >> _______________________________________________ > >> discuss mailing list discuss(a)mpich.org > >> To manage subscription options or unsubscribe: > >> https://lists.mpich.org/mailman/listinfo/discuss > > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > -- Regards Ansar Javed HPC Lab SEECS NUST Contact: +92 334 438 9394 Email: muhammad.ansar(a)seecs.edu.pk

1 0

Re: [mpich-discuss] Running mpi from bash script
by Francisco Pastor 17 Apr '14

17 Apr '14

Hi I've installed a new MPICH version but the problem persists. I looked at mpich.org download page, the version por Ubuntu 12.04 is mpich2-1.4.1 The problem still remains when running from batch. If I run my script from command line everything works fine, if I launch the job with at no processes appear on the cluster nodes and no output files are produced. In master node some processes appear: ps -ef | grep rams meteo 8288 7559 0 13:03 ? 00:00:00 time /usr/bin/mpirun -n 40 -f machinefile ./rams60 -f RAMSIN.operatiu meteo 8289 8288 0 13:03 ? 00:00:00 /usr/bin/mpirun -n 40 -f machinefile ./rams60 -f RAMSIN.operatiu meteo 8295 8290 0 13:03 ? 00:00:00 ./rams60 -f RAMSIN.operatiu meteo 8296 8290 0 13:03 ? 00:00:00 ./rams60 -f RAMSIN.operatiu meteo 8297 8290 0 13:03 ? 00:00:00 ./rams60 -f RAMSIN.operatiu meteo 8298 8290 0 13:03 ? 00:00:00 ./rams60 -f RAMSIN.operatiu meteo 8299 8290 0 13:03 ? 00:00:00 ./rams60 -f RAMSIN.operatiu meteo 8300 8290 0 13:03 ? 00:00:00 ./rams60 -f RAMSIN.operatiu meteo 8301 8290 0 13:03 ? 00:00:00 ./rams60 -f RAMSIN.operatiu meteo 8302 8290 0 13:03 ? 00:00:00 ./rams60 -f RAMSIN.operatiu Should I set some environment when launching the job? Thanks Paco El 17/04/2014 11:06, Reuti escribiï¿½: > Hi, > > Am 17.04.2014 um 10:04 schrieb Francisco Pastor: > >> I need to daily run an mpi job from a bash script. I have prepared the script and run from the command line, this works fine. But when I try to execute at a later time with at (at -f my_script now +1 minute) the parallel processes do not start. >> >> These are the significant parts of my script: >> >> # path to mpi executables >> ruta_mpi=/usr/local/mpich2-1.0.5p4/bin > This version of MPICH is quite old. Please use a more recent one, where the necessity for starting an MPD beforehand does no longer exist. > > -- Reuti > > >> # Function that checks if mpd processes are ok (checks mpdtrace output, five nodes should be up) >> Comprueba_MPD() >> { >> mpdtrace > log.mpd >> wc -l log.mpd > check.mpd >> check1=`awk '{ print $1 }' check.mpd` >> if [ $check1 -eq 5 ]; >> then >> echo "Salida de mpdtrace correcta, hosts=$check1" >> else >> echo "Salida de mpdtrace incorrecta, hosts=$check1" >> $ruta_mpi/mpdboot -n 5 & >> sleep 10 >> mpdtrace >> fi >> } >> >> >> # Run rams60 program with mpirun >> time $ruta_mpi/mpirun -n 21 ./rams60 -f RAMSIN.operatiu >> >> I am sure missing something but I can't find out what's going on. >> >> Thanks in advance >> >> -- >> ----------- >> Dr. Francisco Pastor >> Meteorology department, Instituto Universitario CEAM-UMH >> http://www.ceam.es >> ----------- >> Mendeley profile: http://www.mendeley.com/profiles/francisco-pastor1/ >> Google Scholar: http://scholar.google.com/citations?user=V3mmCdkAAAAJ&hl=es >> Researcher ID: http://www.researcherid.com/rid/B-8331-2008 >> Cosis profile: http://www.cosis.net/profile/francisco.pastor >> ----------- >> mail: paco(a)ceam.es >> skype: paco.pastor.guzman >> ----------- >> Parque Tecnologico, C/ Charles R. Darwin, 14 >> 46980 PATERNA (Valencia), Spain >> Tlf. 96 131 82 27 - Fax. 96 131 81 90 >> >> >> --------------------------------------------------------------------- >> Este mensaje y los ficheros anexos son confidenciales. Los mismos contienen informaciï¿½n reservada de la empresa que no puede ser difundida. Si usted ha recibido este correo por error, tenga la amabilidad de eliminarlo de su sistema y avisar al remitente mediante reenvï¿½o a su direcciï¿½n electrï¿½nica; no deberï¿½ copiar el mensaje ni divulgar su contenido a ninguna persona. >> >> Su direcciï¿½n de correo electrï¿½nico junto a sus datos personales forman parte de un fichero titularidad de la Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213, cuya finalidad es la de mantener el contacto con Ud. De acuerdo con la Ley Orgï¿½nica 15/1999, usted puede ejercitar sus derechos de acceso, rectificaciï¿½n, cancelaciï¿½n y, en su caso, oposiciï¿½n enviando una solicitud por escrito, acompaï¿½ada de una fotocopia de su DNI dirigida a: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia). >> >> This message and the attached files are confidential. They contain reserved information belonging to our centre and are not to be broadcast. If you have received this email by mistake, please delete it from your system and alert the sender by returning it to his/her email address. You must not copy or divulge the contents of the message to anyone. >> >> Your email address and personal data are included in a file belonging to the Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213. The purpose of this file is to allow us to keep in contact with you. In accordance with Organic Law 15/1999, you are permitted to access, rectify, cancel or oppose the contents of this file by submitting a written request, accompanied by a photocopy of your DNI, to: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia). >> >> >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss >> > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > > -- ----------- Dr. Francisco Pastor Meteorology department, Instituto Universitario CEAM-UMH http://www.ceam.es ----------- Mendeley profile: http://www.mendeley.com/profiles/francisco-pastor1/ Google Scholar: http://scholar.google.com/citations?user=V3mmCdkAAAAJ&hl=es Researcher ID: http://www.researcherid.com/rid/B-8331-2008 Cosis profile: http://www.cosis.net/profile/francisco.pastor ----------- mail: paco(a)ceam.es skype: paco.pastor.guzman ----------- Parque Tecnologico, C/ Charles R. Darwin, 14 46980 PATERNA (Valencia), Spain Tlf. 96 131 82 27 - Fax. 96 131 81 90 --------------------------------------------------------------------- Este mensaje y los ficheros anexos son confidenciales. Los mismos contienen informaciï¿½n reservada de la empresa que no puede ser difundida. Si usted ha recibido este correo por error, tenga la amabilidad de eliminarlo de su sistema y avisar al remitente mediante reenvï¿½o a su direcciï¿½n electrï¿½nica; no deberï¿½ copiar el mensaje ni divulgar su contenido a ninguna persona. Su direcciï¿½n de correo electrï¿½nico junto a sus datos personales forman parte de un fichero titularidad de la Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213, cuya finalidad es la de mantener el contacto con Ud. De acuerdo con la Ley Orgï¿½nica 15/1999, usted puede ejercitar sus derechos de acceso, rectificaciï¿½n, cancelaciï¿½n y, en su caso, oposiciï¿½n enviando una solicitud por escrito, acompaï¿½ada de una fotocopia de su DNI dirigida a: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia). This message and the attached files are confidential. They contain reserved information belonging to our centre and are not to be broadcast. If you have received this email by mistake, please delete it from your system and alert the sender by returning it to his/her email address. You must not copy or divulge the contents of the message to anyone. Your email address and personal data are included in a file belonging to the Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213. The purpose of this file is to allow us to keep in contact with you. In accordance with Organic Law 15/1999, you are permitted to access, rectify, cancel or oppose the contents of this file by submitting a written request, accompanied by a photocopy of your DNI, to: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia).

1 0

Re: [mpich-discuss] Running mpi from bash script
by Francisco Pastor 17 Apr '14

17 Apr '14

Hi Reuti The point is that I am using a program, meteorological model RAMS, that was running fine with this mpich2 version. And it runs fine from the command line. If the system works, should I upgrade? Will it work with my program? Can I install two different versions at the same time or should I delete the older one? Is there any chance to run Thanks El 17/04/2014 11:06, Reuti escribiï¿½: > Hi, > > Am 17.04.2014 um 10:04 schrieb Francisco Pastor: > >> I need to daily run an mpi job from a bash script. I have prepared the script and run from the command line, this works fine. But when I try to execute at a later time with at (at -f my_script now +1 minute) the parallel processes do not start. >> >> These are the significant parts of my script: >> >> # path to mpi executables >> ruta_mpi=/usr/local/mpich2-1.0.5p4/bin > This version of MPICH is quite old. Please use a more recent one, where the necessity for starting an MPD beforehand does no longer exist. > > -- Reuti > > >> # Function that checks if mpd processes are ok (checks mpdtrace output, five nodes should be up) >> Comprueba_MPD() >> { >> mpdtrace > log.mpd >> wc -l log.mpd > check.mpd >> check1=`awk '{ print $1 }' check.mpd` >> if [ $check1 -eq 5 ]; >> then >> echo "Salida de mpdtrace correcta, hosts=$check1" >> else >> echo "Salida de mpdtrace incorrecta, hosts=$check1" >> $ruta_mpi/mpdboot -n 5 & >> sleep 10 >> mpdtrace >> fi >> } >> >> >> # Run rams60 program with mpirun >> time $ruta_mpi/mpirun -n 21 ./rams60 -f RAMSIN.operatiu >> >> I am sure missing something but I can't find out what's going on. >> >> Thanks in advance >> >> -- >> ----------- >> Dr. Francisco Pastor >> Meteorology department, Instituto Universitario CEAM-UMH >> http://www.ceam.es >> ----------- >> Mendeley profile: http://www.mendeley.com/profiles/francisco-pastor1/ >> Google Scholar: http://scholar.google.com/citations?user=V3mmCdkAAAAJ&hl=es >> Researcher ID: http://www.researcherid.com/rid/B-8331-2008 >> Cosis profile: http://www.cosis.net/profile/francisco.pastor >> ----------- >> mail: paco(a)ceam.es >> skype: paco.pastor.guzman >> ----------- >> Parque Tecnologico, C/ Charles R. Darwin, 14 >> 46980 PATERNA (Valencia), Spain >> Tlf. 96 131 82 27 - Fax. 96 131 81 90 >> >> >> --------------------------------------------------------------------- >> Este mensaje y los ficheros anexos son confidenciales. Los mismos contienen informaciï¿½n reservada de la empresa que no puede ser difundida. Si usted ha recibido este correo por error, tenga la amabilidad de eliminarlo de su sistema y avisar al remitente mediante reenvï¿½o a su direcciï¿½n electrï¿½nica; no deberï¿½ copiar el mensaje ni divulgar su contenido a ninguna persona. >> >> Su direcciï¿½n de correo electrï¿½nico junto a sus datos personales forman parte de un fichero titularidad de la Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213, cuya finalidad es la de mantener el contacto con Ud. De acuerdo con la Ley Orgï¿½nica 15/1999, usted puede ejercitar sus derechos de acceso, rectificaciï¿½n, cancelaciï¿½n y, en su caso, oposiciï¿½n enviando una solicitud por escrito, acompaï¿½ada de una fotocopia de su DNI dirigida a: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia). >> >> This message and the attached files are confidential. They contain reserved information belonging to our centre and are not to be broadcast. If you have received this email by mistake, please delete it from your system and alert the sender by returning it to his/her email address. You must not copy or divulge the contents of the message to anyone. >> >> Your email address and personal data are included in a file belonging to the Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213. The purpose of this file is to allow us to keep in contact with you. In accordance with Organic Law 15/1999, you are permitted to access, rectify, cancel or oppose the contents of this file by submitting a written request, accompanied by a photocopy of your DNI, to: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia). >> >> >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss >> > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > > -- ----------- Dr. Francisco Pastor Meteorology department, Instituto Universitario CEAM-UMH http://www.ceam.es ----------- Mendeley profile: http://www.mendeley.com/profiles/francisco-pastor1/ Google Scholar: http://scholar.google.com/citations?user=V3mmCdkAAAAJ&hl=es Researcher ID: http://www.researcherid.com/rid/B-8331-2008 Cosis profile: http://www.cosis.net/profile/francisco.pastor ----------- mail: paco(a)ceam.es skype: paco.pastor.guzman ----------- Parque Tecnologico, C/ Charles R. Darwin, 14 46980 PATERNA (Valencia), Spain Tlf. 96 131 82 27 - Fax. 96 131 81 90 --------------------------------------------------------------------- Este mensaje y los ficheros anexos son confidenciales. Los mismos contienen informaciï¿½n reservada de la empresa que no puede ser difundida. Si usted ha recibido este correo por error, tenga la amabilidad de eliminarlo de su sistema y avisar al remitente mediante reenvï¿½o a su direcciï¿½n electrï¿½nica; no deberï¿½ copiar el mensaje ni divulgar su contenido a ninguna persona. Su direcciï¿½n de correo electrï¿½nico junto a sus datos personales forman parte de un fichero titularidad de la Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213, cuya finalidad es la de mantener el contacto con Ud. De acuerdo con la Ley Orgï¿½nica 15/1999, usted puede ejercitar sus derechos de acceso, rectificaciï¿½n, cancelaciï¿½n y, en su caso, oposiciï¿½n enviando una solicitud por escrito, acompaï¿½ada de una fotocopia de su DNI dirigida a: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia). This message and the attached files are confidential. They contain reserved information belonging to our centre and are not to be broadcast. If you have received this email by mistake, please delete it from your system and alert the sender by returning it to his/her email address. You must not copy or divulge the contents of the message to anyone. Your email address and personal data are included in a file belonging to the Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213. The purpose of this file is to allow us to keep in contact with you. In accordance with Organic Law 15/1999, you are permitted to access, rectify, cancel or oppose the contents of this file by submitting a written request, accompanied by a photocopy of your DNI, to: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia).

1 0

Re: [mpich-discuss] Running mpi from bash script
by Reuti 17 Apr '14

17 Apr '14

Hi, Am 17.04.2014 um 10:04 schrieb Francisco Pastor: > I need to daily run an mpi job from a bash script. I have prepared the script and run from the command line, this works fine. But when I try to execute at a later time with at (at -f my_script now +1 minute) the parallel processes do not start. > > These are the significant parts of my script: > > # path to mpi executables > ruta_mpi=/usr/local/mpich2-1.0.5p4/bin This version of MPICH is quite old. Please use a more recent one, where the necessity for starting an MPD beforehand does no longer exist. -- Reuti > # Function that checks if mpd processes are ok (checks mpdtrace output, five nodes should be up) > Comprueba_MPD() > { > mpdtrace > log.mpd > wc -l log.mpd > check.mpd > check1=`awk '{ print $1 }' check.mpd` > if [ $check1 -eq 5 ]; > then > echo "Salida de mpdtrace correcta, hosts=$check1" > else > echo "Salida de mpdtrace incorrecta, hosts=$check1" > $ruta_mpi/mpdboot -n 5 & > sleep 10 > mpdtrace > fi > } > > > # Run rams60 program with mpirun > time $ruta_mpi/mpirun -n 21 ./rams60 -f RAMSIN.operatiu > > I am sure missing something but I can't find out what's going on. > > Thanks in advance > > -- > ----------- > Dr. Francisco Pastor > Meteorology department, Instituto Universitario CEAM-UMH > http://www.ceam.es > ----------- > Mendeley profile: http://www.mendeley.com/profiles/francisco-pastor1/ > Google Scholar: http://scholar.google.com/citations?user=V3mmCdkAAAAJ&hl=es > Researcher ID: http://www.researcherid.com/rid/B-8331-2008 > Cosis profile: http://www.cosis.net/profile/francisco.pastor > ----------- > mail: paco(a)ceam.es > skype: paco.pastor.guzman > ----------- > Parque Tecnologico, C/ Charles R. Darwin, 14 > 46980 PATERNA (Valencia), Spain > Tlf. 96 131 82 27 - Fax. 96 131 81 90 > > > --------------------------------------------------------------------- > Este mensaje y los ficheros anexos son confidenciales. Los mismos contienen información reservada de la empresa que no puede ser difundida. Si usted ha recibido este correo por error, tenga la amabilidad de eliminarlo de su sistema y avisar al remitente mediante reenvío a su dirección electrónica; no deberá copiar el mensaje ni divulgar su contenido a ninguna persona. > > Su dirección de correo electrónico junto a sus datos personales forman parte de un fichero titularidad de la Fundación de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterráneo - CEAM, con CIF: G-46957213, cuya finalidad es la de mantener el contacto con Ud. De acuerdo con la Ley Orgánica 15/1999, usted puede ejercitar sus derechos de acceso, rectificación, cancelación y, en su caso, oposición enviando una solicitud por escrito, acompañada de una fotocopia de su DNI dirigida a: Fundación de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterráneo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnológico.46980 PATERNA (Valencia). > > This message and the attached files are confidential. They contain reserved information belonging to our centre and are not to be broadcast. If you have received this email by mistake, please delete it from your system and alert the sender by returning it to his/her email address. You must not copy or divulge the contents of the message to anyone. > > Your email address and personal data are included in a file belonging to the Fundación de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterráneo - CEAM, con CIF: G-46957213. The purpose of this file is to allow us to keep in contact with you. In accordance with Organic Law 15/1999, you are permitted to access, rectify, cancel or oppose the contents of this file by submitting a written request, accompanied by a photocopy of your DNI, to: Fundación de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterráneo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnológico.46980 PATERNA (Valencia). > > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss >

1 0

Running mpi from bash script
by Francisco Pastor 17 Apr '14

17 Apr '14

Hi everyone I need to daily run an mpi job from a bash script. I have prepared the script and run from the command line, this works fine. But when I try to execute at a later time with at (at -f my_script now +1 minute) the parallel processes do not start. These are the significant parts of my script: # path to mpi executables ruta_mpi=/usr/local/mpich2-1.0.5p4/bin # Function that checks if mpd processes are ok (checks mpdtrace output, five nodes should be up) Comprueba_MPD() { mpdtrace > log.mpd wc -l log.mpd > check.mpd check1=`awk '{ print $1 }' check.mpd` if [ $check1 -eq 5 ]; then echo "Salida de mpdtrace correcta, hosts=$check1" else echo "Salida de mpdtrace incorrecta, hosts=$check1" $ruta_mpi/mpdboot -n 5 & sleep 10 mpdtrace fi } # Run rams60 program with mpirun time $ruta_mpi/mpirun -n 21 ./rams60 -f RAMSIN.operatiu I am sure missing something but I can't find out what's going on. Thanks in advance -- ----------- Dr. Francisco Pastor Meteorology department, Instituto Universitario CEAM-UMH http://www.ceam.es ----------- Mendeley profile: http://www.mendeley.com/profiles/francisco-pastor1/ Google Scholar: http://scholar.google.com/citations?user=V3mmCdkAAAAJ&hl=es Researcher ID: http://www.researcherid.com/rid/B-8331-2008 Cosis profile: http://www.cosis.net/profile/francisco.pastor ----------- mail: paco(a)ceam.es skype: paco.pastor.guzman ----------- Parque Tecnologico, C/ Charles R. Darwin, 14 46980 PATERNA (Valencia), Spain Tlf. 96 131 82 27 - Fax. 96 131 81 90 --------------------------------------------------------------------- Este mensaje y los ficheros anexos son confidenciales. Los mismos contienen informaciï¿½n reservada de la empresa que no puede ser difundida. Si usted ha recibido este correo por error, tenga la amabilidad de eliminarlo de su sistema y avisar al remitente mediante reenvï¿½o a su direcciï¿½n electrï¿½nica; no deberï¿½ copiar el mensaje ni divulgar su contenido a ninguna persona. Su direcciï¿½n de correo electrï¿½nico junto a sus datos personales forman parte de un fichero titularidad de la Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213, cuya finalidad es la de mantener el contacto con Ud. De acuerdo con la Ley Orgï¿½nica 15/1999, usted puede ejercitar sus derechos de acceso, rectificaciï¿½n, cancelaciï¿½n y, en su caso, oposiciï¿½n enviando una solicitud por escrito, acompaï¿½ada de una fotocopia de su DNI dirigida a: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia). This message and the attached files are confidential. They contain reserved information belonging to our centre and are not to be broadcast. If you have received this email by mistake, please delete it from your system and alert the sender by returning it to his/her email address. You must not copy or divulge the contents of the message to anyone. Your email address and personal data are included in a file belonging to the Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM, con CIF: G-46957213. The purpose of this file is to allow us to keep in contact with you. In accordance with Organic Law 15/1999, you are permitted to access, rectify, cancel or oppose the contents of this file by submitting a written request, accompanied by a photocopy of your DNI, to: Fundaciï¿½n de la Comunidad Valenciana Centro de Estudios Ambientales del Mediterrï¿½neo - CEAM. C/ Charles R. Darwin, 14. Parque Tecnolï¿½gico.46980 PATERNA (Valencia).

1 0

MPJ Express under MPICH
by Bibrak Qamar 17 Apr '14

17 Apr '14

Dear MPICH Community, We are please to announce the next beta release of MPJ Express (0.41). This release adds supports for using a native MPI library like MPICH and its derivatives (MVAPICH and MS-MPI). This helps run parallel Java applications on top of MPICH. Please try it out http://mpjexpress.org/ and let us know if you face issues :) regards, Bibrak Qamar

1 0

Re: [mpich-discuss] Is it a correct program? Sending without receiving.
by Lu, Huiwei 16 Apr '14

16 Apr '14

Thanks Wesley. — Huiwei On Apr 16, 2014, at 4:48 PM, Wesley Bland <wbland(a)anl.gov> wrote: > It's the user's responsibility. The only guarantee made by MPI_Send is that the data can be overwritten by the user. This may mean that it's copied into a buffer locally and will be sent later. If the data is being sent eagerly, this is exactly what's happened. If you want to make sure the data is actually received, you need to use MPI_Ssend. > > > On Wed, Apr 16, 2014 at 4:36 PM, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > One of our group members is using MPI_Send in his program without receiving. I am surprised why his program will finish. > > Here is the program: > > #include <mpi.h> > #include <stdio.h> > int main(int argc,char** argv) > { > int rank; > int size; > MPI_Init(&argc, &argv); > MPI_Comm_rank(MPI_COMM_WORLD,&rank); > MPI_Comm_size(MPI_COMM_WORLD,&size); > > int aa = 10; > if(rank == 0) { > MPI_Send(&aa, 1, MPI_INT, 1, 0,MPI_COMM_WORLD); > } > MPI_Barrier(MPI_COMM_WORLD); > printf("%d, complete barrier!\n",rank); > > MPI_Finalize(); > } > > When I run the above program with: mpiexec -n 2 ./mpi_send. The output is: > 0, complete barrier! > 1, complete barrier! > In direct memory block for handle type REQUEST, 1 handles are still allocated > [1] 8 at [0x00007fe79a7026b8], src/mpid/ch3/src/ch3u_eager.c[440] > > The warning tells me the send message is sent using eager protocol. Is it right behavior for an MPI_Send to return without actually delivering the data? > Or is it user’s responsibility to pair all the send/receive calls. > > — > Huiwei Lu > Postdoc Appointee > Mathematics and Computer Science Division, Argonne National Laboratory > http://www.mcs.anl.gov/~huiweilu/ > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] Is it a correct program? Sending without receiving.
by Wesley Bland 16 Apr '14

16 Apr '14

It's the user's responsibility. The only guarantee made by MPI_Send is that the data can be overwritten by the user. This may mean that it's copied into a buffer locally and will be sent later. If the data is being sent eagerly, this is exactly what's happened. If you want to make sure the data is actually received, you need to use MPI_Ssend. On Wed, Apr 16, 2014 at 4:36 PM, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > One of our group members is using MPI_Send in his program without > receiving. I am surprised why his program will finish. > > Here is the program: > > #include <mpi.h> > #include <stdio.h> > int main(int argc,char** argv) > { > int rank; > int size; > MPI_Init(&argc, &argv); > MPI_Comm_rank(MPI_COMM_WORLD,&rank); > MPI_Comm_size(MPI_COMM_WORLD,&size); > > int aa = 10; > if(rank == 0) { > MPI_Send(&aa, 1, MPI_INT, 1, 0,MPI_COMM_WORLD); > } > MPI_Barrier(MPI_COMM_WORLD); > printf("%d, complete barrier!\n",rank); > > MPI_Finalize(); > } > > When I run the above program with: mpiexec -n 2 ./mpi_send. The output is: > 0, complete barrier! > 1, complete barrier! > In direct memory block for handle type REQUEST, 1 handles are still > allocated > [1] 8 at [0x00007fe79a7026b8], src/mpid/ch3/src/ch3u_eager.c[440] > > The warning tells me the send message is sent using eager protocol. Is it > right behavior for an MPI_Send to return without actually delivering the > data? > Or is it user’s responsibility to pair all the send/receive calls. > > — > Huiwei Lu > Postdoc Appointee > Mathematics and Computer Science Division, Argonne National Laboratory > http://www.mcs.anl.gov/~huiweilu/ > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss >

1 0

Is it a correct program? Sending without receiving.
by Lu, Huiwei 16 Apr '14

16 Apr '14

One of our group members is using MPI_Send in his program without receiving. I am surprised why his program will finish. Here is the program: #include <mpi.h> #include <stdio.h> int main(int argc,char** argv) { int rank; int size; MPI_Init(&argc, &argv); MPI_Comm_rank(MPI_COMM_WORLD,&rank); MPI_Comm_size(MPI_COMM_WORLD,&size); int aa = 10; if(rank == 0) { MPI_Send(&aa, 1, MPI_INT, 1, 0,MPI_COMM_WORLD); } MPI_Barrier(MPI_COMM_WORLD); printf("%d, complete barrier!\n",rank); MPI_Finalize(); } When I run the above program with: mpiexec -n 2 ./mpi_send. The output is: 0, complete barrier! 1, complete barrier! In direct memory block for handle type REQUEST, 1 handles are still allocated [1] 8 at [0x00007fe79a7026b8], src/mpid/ch3/src/ch3u_eager.c[440] The warning tells me the send message is sent using eager protocol. Is it right behavior for an MPI_Send to return without actually delivering the data? Or is it user’s responsibility to pair all the send/receive calls. — Huiwei Lu Postdoc Appointee Mathematics and Computer Science Division, Argonne National Laboratory http://www.mcs.anl.gov/~huiweilu/

1 0

Compression in ROMIO
by Sanjuan, Alejandro 16 Apr '14

16 Apr '14

Good Afternoon, I am working in a research about compression in ROMIO and I have a few problems for compressing/decompressing the noncontiguous types. I have implemented a compression algorithm in the functions ADIOI_GEN_ReadContig/ADIOI_GEN_WriteContig that works fine (in UFS). For the noncontiguous case I have tried to let it like this given that ADIOI_GEN_ReadStrided/ADIOI_GEN_WriteStrided calls to the functions modified before, but this is not working. Certainly I cannot understand some pieces of the code, and I cannot understand what is going on. My question is, how does the read-modify-write operation work (which I suppose is implemented in the macro ADIOI_BUFFERED_WRITE)? What is your opinion about the possibility/impossibility of implementing compression this way? Regards, Alejandro Sanjuan

1 0