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

Re: [mpich-discuss] Does MPICH support launching processes on Intel-MIC from host?
by Min Si 23 May '14

23 May '14

Have you tried ssh to MIC and check mpiexec.hydra again ? It might be different files just using the same path. Min (14-5-23 15:48), Yida Wang wrote: > I don't know how Intel guys implement this. But I think the > mpiexec.hydra file they provide is a regular Linux executable file. > "file" it returns "mpiexec.hydra: ELF 64-bit LSB executable, x86-64, > version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux > 2.6.9, not stripped". > > One need to make sure mpiexec.hydra and pmi_proxy are accessible to > MICs before launching processes on MICs from host. > > > On Fri, May 23, 2014 at 4:43 PM, Reuti <reuti(a)staff.uni-marburg.de > <mailto:[email protected]>> wrote: > > Am 23.05.2014 um 22:38 schrieb Yida Wang: > >> Thank you! That works, but is not very convenient. Can I know why >> you guys don't configure mpich to behave as the same as >> Intel-MPI? Basically, using Intel-MPI, one can use the same >> mpiexec.hydra on both host and Xeon Phi. > > I don't have Intel-MPI, but is it really a binary or a wrapper to > two different versions? What does `file mpiexec.hydra` show? > > -- Reuti > > >> On Fri, May 23, 2014 at 1:18 PM, "Antonio J. Peï¿½a" >> <apenya(a)mcs.anl.gov <mailto:[email protected]>> wrote: >> >> >> For that, you have to deploy two installations of MPICH: one >> compiled for the host, and the other for the MIC. To make >> things easy, place them under the same path in both hosts - >> one that is not shared. >> >> Antonio >> >> >> >> On 05/22/2014 04:28 PM, Yida Wang wrote: >>> Hi Guys, >>> >>> Can MPICH be configured to support launching processes on >>> Intel-MIC from the host? I know that there's a way to >>> compile MPICH to be running on Intel-MIC only, but how about >>> running heterogeneously on both host and Intel-MIC together? >>> >>> My compiled version returned "hydra_pmi_proxy: cannot >>> execute binary file" and got stuck when I tried to launch >>> processes on Intel-MIC. >>> >>> Thanks, >>> YW >>> >>> >>> _______________________________________________ >>> discuss mailing listdiscuss(a)mpich.org <mailto:[email protected]> >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/discuss >> >> >> -- >> Antonio J. Peï¿½a >> 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 <mailto:[email protected]> >> www.mcs.anl.gov/~apenya <http://www.mcs.anl.gov/%7Eapenya> >> >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org <mailto:[email protected]> >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss >> >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org <mailto:[email protected]> >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org <mailto:[email protected]> > 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] Does MPICH support launching processes on Intel-MIC from host?
by Yida Wang 23 May '14

23 May '14

I don't know how Intel guys implement this. But I think the mpiexec.hydra file they provide is a regular Linux executable file. "file" it returns "mpiexec.hydra: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.9, not stripped". One need to make sure mpiexec.hydra and pmi_proxy are accessible to MICs before launching processes on MICs from host. On Fri, May 23, 2014 at 4:43 PM, Reuti <reuti(a)staff.uni-marburg.de> wrote: > Am 23.05.2014 um 22:38 schrieb Yida Wang: > > Thank you! That works, but is not very convenient. Can I know why you guys > don't configure mpich to behave as the same as Intel-MPI? Basically, using > Intel-MPI, one can use the same mpiexec.hydra on both host and Xeon Phi. > > > I don't have Intel-MPI, but is it really a binary or a wrapper to two > different versions? What does `file mpiexec.hydra` show? > > -- Reuti > > > On Fri, May 23, 2014 at 1:18 PM, "Antonio J. Peña" <apenya(a)mcs.anl.gov>wrote: > >> >> For that, you have to deploy two installations of MPICH: one compiled for >> the host, and the other for the MIC. To make things easy, place them under >> the same path in both hosts - one that is not shared. >> >> Antonio >> >> >> >> On 05/22/2014 04:28 PM, Yida Wang wrote: >> >> Hi Guys, >> >> Can MPICH be configured to support launching processes on Intel-MIC >> from the host? I know that there's a way to compile MPICH to be running on >> Intel-MIC only, but how about running heterogeneously on both host and >> Intel-MIC together? >> >> My compiled version returned "hydra_pmi_proxy: cannot execute binary >> file" and got stuck when I tried to launch processes on Intel-MIC. >> >> Thanks, >> YW >> >> >> _______________________________________________ >> 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 60439-4847apenya(a)mcs.anl.govwww.mcs.anl.gov/~apenya >> >> >> _______________________________________________ >> 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 > > > > _______________________________________________ > 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] Does MPICH support launching processes on Intel-MIC from host?
by Reuti 23 May '14

23 May '14

Am 23.05.2014 um 22:38 schrieb Yida Wang: > Thank you! That works, but is not very convenient. Can I know why you guys don't configure mpich to behave as the same as Intel-MPI? Basically, using Intel-MPI, one can use the same mpiexec.hydra on both host and Xeon Phi. I don't have Intel-MPI, but is it really a binary or a wrapper to two different versions? What does `file mpiexec.hydra` show? -- Reuti > On Fri, May 23, 2014 at 1:18 PM, "Antonio J. Peña" <apenya(a)mcs.anl.gov> wrote: > > For that, you have to deploy two installations of MPICH: one compiled for the host, and the other for the MIC. To make things easy, place them under the same path in both hosts - one that is not shared. > > Antonio > > > > On 05/22/2014 04:28 PM, Yida Wang wrote: >> Hi Guys, >> >> Can MPICH be configured to support launching processes on Intel-MIC from the host? I know that there's a way to compile MPICH to be running on Intel-MIC only, but how about running heterogeneously on both host and Intel-MIC together? >> >> My compiled version returned "hydra_pmi_proxy: cannot execute binary file" and got stuck when I tried to launch processes on Intel-MIC. >> >> Thanks, >> YW >> >> >> _______________________________________________ >> 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 > > _______________________________________________ > 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] Does MPICH support launching processes on Intel-MIC from host?
by Yida Wang 23 May '14

23 May '14

Thank you! That works, but is not very convenient. Can I know why you guys don't configure mpich to behave as the same as Intel-MPI? Basically, using Intel-MPI, one can use the same mpiexec.hydra on both host and Xeon Phi. On Fri, May 23, 2014 at 1:18 PM, "Antonio J. Peña" <apenya(a)mcs.anl.gov>wrote: > > For that, you have to deploy two installations of MPICH: one compiled for > the host, and the other for the MIC. To make things easy, place them under > the same path in both hosts - one that is not shared. > > Antonio > > > > On 05/22/2014 04:28 PM, Yida Wang wrote: > > Hi Guys, > > Can MPICH be configured to support launching processes on Intel-MIC from > the host? I know that there's a way to compile MPICH to be running on > Intel-MIC only, but how about running heterogeneously on both host and > Intel-MIC together? > > My compiled version returned "hydra_pmi_proxy: cannot execute binary > file" and got stuck when I tried to launch processes on Intel-MIC. > > Thanks, > YW > > > _______________________________________________ > 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 60439-4847apenya(a)mcs.anl.govwww.mcs.anl.gov/~apenya > > > _______________________________________________ > 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] Fwd: Make MPICH on Arch Linux
by Scott O'Malley 23 May '14

23 May '14

It's a Raspberry Pi, I'm building a cluster as part of my MSc project. I'll try build elsewhere and report back Cheers Scott On 23 May 2014 19:05, Kenneth Raffenetti <raffenet(a)mcs.anl.gov> wrote: > I think you may be running into the OOM Killer. Is memory usage high on > your machine? Is /tmp a tmpfs and not a real partition? If so, can you try > building elsewhere? > > Ken > > > On 05/23/2014 12:54 PM, Scott O'Malley wrote: > >> Hi Guys >> >> I'm trying to build MPICH for Arch but when I get to the Make section of >> the user guide 3.1 I get an error, I've looked through the m.txt file as >> the guide suggests but the error message isn't particularly helpful. >> >> >> >> -- >> - Scott >> >> >> >> -- >> - Scott >> >> >> _______________________________________________ >> 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 > -- - Scott

1 0

Re: [mpich-discuss] Fwd: Make MPICH on Arch Linux
by Kenneth Raffenetti 23 May '14

23 May '14

I think you may be running into the OOM Killer. Is memory usage high on your machine? Is /tmp a tmpfs and not a real partition? If so, can you try building elsewhere? Ken On 05/23/2014 12:54 PM, Scott O'Malley wrote: > Hi Guys > > I'm trying to build MPICH for Arch but when I get to the Make section of > the user guide 3.1 I get an error, I've looked through the m.txt file as > the guide suggests but the error message isn't particularly helpful. > > > > -- > - Scott > > > > -- > - Scott > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss >

1 0

Fwd: Make MPICH on Arch Linux
by Scott O'Malley 23 May '14

23 May '14

Hi Guys I'm trying to build MPICH for Arch but when I get to the Make section of the user guide 3.1 I get an error, I've looked through the m.txt file as the guide suggests but the error message isn't particularly helpful. -- - Scott -- - Scott

1 0

Re: [mpich-discuss] Error: "MPIR_Get_contextid_sparse_group(1193): Too many communicators (0/2048)"
by Rajeev Thakur 23 May '14

23 May '14

I don't fully understand your use case, but: The window memory passed to MPI_Win_create on each process must be a contiguous memory. You can internally manage it in whatever way. If you create a datatype on a process, the datatype is local to that process. Other processes can't use that datatype. Also the MPI standard says the following about the target datatype passed to put/get/accumulate (MPI-3, pg 420, line 33): "The target datatype must contain only relative displacements, not absolute addresses." Rajeev On May 23, 2014, at 9:44 AM, Ted Sariyski <tsariysk(a)craft-tech.com> wrote: > I need a little bit more help. My question relates to the structure of the array (referred below as A), for which a window object will be created. > > Let say that particles are of type particle_t: > > type particle_t > integer :: host, id,... > real,pointer :: data(:)=>null() > end type particle_t > > I built two corresponding mpi types, one with data allocated, mpi_particles_t, and mpi_particles_null_t, when p%data=>null(). > > What I want to do is the following: > • All processes allocates an array of type particles_t for all particles, but data is allocated only if a particle is assigned to the domain: > allocate(A(tot_num_particles)) > do n=1,tot_num_particles > if ( particle(n)%host == myid) allocate(particle(n)%data(data_size)) > enddo > > 2. All processes create their mpi data type mpi_A_t: > > do n=1,tot_num_particles > > if ( particle(n)%host == myid) then > blockcounts(n) = 1 ; oldtypes(n) = mpi_particles_t > else > blockcounts(n) = 1 ; oldtypes(n) = mpi_particles_null_t > endif > call MPI_GET_ADDRESS(A(n), offsets(n), ierr) > enddo > > call mpi_type_create_struct(tot_num_particles,blockcounts,offsets,oldtypes,mpi_A_myid_t,ierr) > call mpi_type_commit(mpi_A_myid_t,ierr) > > I gave a try but it fails. Before I proceed I want to make sure that there is nothing fundamentally wrong with this approach. Besides, even if it is correct, I am not sure that this is the 'best' solution. I will highly appreciate your comments. > > Thanks in advance, > --Ted > > > > > > On 05/22/2014 03:31 PM, Ted Sariyski wrote: >> I see. Thanks a lot. >> --Ted >> >> >> On 05/22/2014 03:15 PM, Rajeev Thakur wrote: >>>> What do you mean with: "Why can’t all processes open one large window?" I guess I miss something. >>> If each process has an array of objects that belong to it (called array A say) , then with a single call to MPI_Win_create you can create a window object that has everyone's A arrays in it. >>> >>> Rajeev >>> >>> >>> >>> On May 22, 2014, at 9:46 AM, Ted Sariyski <tsariysk(a)craft-tech.com> wrote: >>> >>>> It is about MPI_win. Here is the problem as it relates to MPI (it is a Boltzmann type equation). >>>> >>>> There are N particles interacting with each other. Interaction is directional, so that a particle interacts only with those particles, which are within a narrow cone. The first step is to segregate the initial set of N particles into subsets of particles (I called it 'objects'), which interact with each other. Here is what I do: >>>> • Assign each object to a process. >>>> • The process which owns a object: >>>> • Makes a guess for the maximum number of particles expected in this object. >>>> • Allocates memory for it. >>>> • Opens a shared window. >>>> • All processes >>>> • Each particle identify which object it belongs to, and PUTs its data there. >>>> • After assembly is done, objects are passed to a solver. >>>> • Repeat >>>> What do you mean with: "Why can’t all processes open one large window?" I guess I miss something. >>>> Thanks, >>>> --Ted >>>> >>>> >>>> On 05/21/2014 11:03 PM, Balaji, Pavan wrote: >>>>> On May 21, 2014, at 6:02 PM, Ted Sariyski <tsariysk(a)craft-tech.com> >>>>> wrote: >>>>> >>>>>> Memory limitations. With one large window all processes have to allocate memory for the objects they own as well as for objects assigned to other process. >>>>>> >>>>> Are we talking about the same thing here? I’m referring to an MPI_Win. What objects do processes need to keep track of? >>>>> >>>>> — Pavan >>>>> >>>>> _______________________________________________ >>>>> 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 >>> _______________________________________________ >>> 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] Does MPICH support launching processes on Intel-MIC from host?
by "Antonio J. Peña" 23 May '14

23 May '14

For that, you have to deploy two installations of MPICH: one compiled for the host, and the other for the MIC. To make things easy, place them under the same path in both hosts - one that is not shared. Antonio On 05/22/2014 04:28 PM, Yida Wang wrote: > Hi Guys, > > Can MPICH be configured to support launching processes on Intel-MIC > from the host? I know that there's a way to compile MPICH to be > running on Intel-MIC only, but how about running heterogeneously on > both host and Intel-MIC together? > > My compiled version returned "hydra_pmi_proxy: cannot execute binary > file" and got stuck when I tried to launch processes on Intel-MIC. > > Thanks, > YW > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Antonio J. Peï¿½a 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

1 0

Re: [mpich-discuss] process failing...
by Rajeev Thakur 23 May '14

23 May '14

Yes. The message below says some process has received 261,895 messages for which no matching receives have been posted yet. Rajeev > It looks like at least one of your processes is receiving too many unexpected messages, leading to get out of memory. Unexpected messages are those not matching a posted receive on the receiver side. You may check with the application developers to make them review the algorithm or look for any possible bug. > > Antonio

1 0