March 2015 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] website seems broken
by "Antonio J. Peña" 03 Mar '15

03 Mar '15

The download links are up again in our website. Antonio On 03/03/2015 11:05 AM, "Antonio J. Peña" wrote: > > Hi Jean-Matthieu, > > We are aware of this issue and are working with our system > administrators to solve it as soon as possible. As an alternative, you > can get the source code directly from our repository: > |http://git.mpich.org/mpich.git| (check out the v3.1.4 tag). I will > send you a notice once the website is back to normal operations. > > Best, > Antonio > > > On 03/03/2015 07:41 AM, Jean-Matthieu Etancelin wrote: >> Hi, >> >> The ‘Download’ page seems broken : http://www.mpich.org/downloads/ >> Download links for latests release are pointing the same page so we >> can’t get sources. Is there an other way ? >> >> I hope it could be fixed soon. >> Best regards >> >> — >> Jean-Matthieu Etancelin >> >> >> _______________________________________________ >> discuss mailing listdiscuss(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 -- 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

1 0

Re: [mpich-discuss] website seems broken
by "Antonio J. Peña" 03 Mar '15

03 Mar '15

Hi Jean-Matthieu, We are aware of this issue and are working with our system administrators to solve it as soon as possible. As an alternative, you can get the source code directly from our repository: |http://git.mpich.org/mpich.git| (check out the v3.1.4 tag). I will send you a notice once the website is back to normal operations. Best, Antonio On 03/03/2015 07:41 AM, Jean-Matthieu Etancelin wrote: > Hi, > > The ‘Download’ page seems broken : http://www.mpich.org/downloads/ > Download links for latests release are pointing the same page so we > can’t get sources. Is there an other way ? > > I hope it could be fixed soon. > Best regards > > — > Jean-Matthieu Etancelin > > > _______________________________________________ > 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

1 0

website seems broken
by Jean-Matthieu Etancelin 03 Mar '15

03 Mar '15

Hi, The ‘Download’ page seems broken : http://www.mpich.org/downloads/ <http://www.mpich.org/downloads/> Download links for latests release are pointing the same page so we can’t get sources. Is there an other way ? I hope it could be fixed soon. Best regards — Jean-Matthieu Etancelin _______________________________________________ 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] MPI scatter algorithms for BlueGene/P
by Khalid Hasanov 03 Mar '15

03 Mar '15

Jeff, thanks a lot for your detailed answer. It makes much more sense now. Best regards, Khalid On Tue, Mar 3, 2015 at 4:11 AM, Jeff Hammond <jeff.science(a)gmail.com> wrote: > Those operations were never specially optimized on Blue Gene. The > operations that were optimized heavily on Blue Gene were/are Barrier, > Bcast, Allreduce, and Alltoall(v). Most of the other collective > optimizations were derivatives of that. For example, one can > implement Allgather as a series of Bcasts. And Reduce was optimized > as a side effect of Allreduce; in some cases, Allreduce was faster > than Reduce, which is counter-intuitive. > > One optimization that Blue Gene had for Gather was to use Reduce with > BOR (on BG/P) and SUM (on BG/Q, for float types at least). This > turned Gather of count=1 into Reduce of count=nproc (size of > communicator), but it was very faster for short messages. > > There aren't many good optimizations for Scatter. MPICH has most of > the generic ones, as you might expect. It is possible that Scatter as > Alltoallv with only one non-zero in the count vector was faster than > MPICH, but this would surprise me. > > In any case, all MPI collectives on BG benefit from very good > network-to-processor balance, minimal rendezvous (connectionless HW > and SW), trivial virtual-to-physical translation, good bisection > bandwidth of _electrically isolated_ torus networks, etc. > > If you have a more specific question, I might have a better answer. > > Best, > > Jeff > > On Sun, Mar 1, 2015 at 4:36 PM, Khalid Hasanov <xalid.h(a)gmail.com> wrote: > > Hello, > > > > First of all, I am not sure if this group is the right place for this > > question. If not I apologize for asking unrelated question. > > > > I read two papers about optimizing MPI collective communications on BG/L > and > > BG/P. > > (Optimization of MPI Collective Communication on BlueGene/L Systems > > and MPI Collective Communications on The Blue Gene/P Supercomputer: > > Algorithms and Optimizations respectively). However, these two papers do > not > > mention anything about MPI scatter and gather operations, I wonder if > these > > two collective operations have been optimized for BlueGene or they use > > exactly the same algorithms from MPICH. Any reference appreciated. > > > > > > Best regards, > > Khalid > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > jeff.science(a)gmail.com > http://jeffhammond.github.io/ > _______________________________________________ > 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

calculated a execution time in a parallel programe
by Chafik sanaa 03 Mar '15

03 Mar '15

I use the MPICH2 to parallelize my algorithm on an i5 machine, is the time calculated in this type of machinery is the time actually consumed or estimated in my programe? I ask this question because someone told me that if you do not work on a parallel machine then even if you calculate the execution time in your parallel algorithm and you compare it with the algorithm that you do not use parallelism is not correct.effectively I tried to compare the execution time of both programs but really I find that there's a accelerartion when I use the parallelism _______________________________________________ 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] MPI_Finalize hangs in dynamic connection in case of failed process
by Roy, Hirak 02 Mar '15

02 Mar '15

Hi Wesley, As I mentioned in my email, MPI_disconnect hangs. Here is a short program which you can run. Please note that there is an "assert" in client.c Compile : >> mpicc server.c -o server >> mpicc client.c -o client To run use two shell/terminal : Term1>> mpiexec -n 1 ./server Term2>> mpiexec -n 1 ./client Please press any key on the server terminal. I earlier filed a bug related to this : http://trac.mpich.org/projects/mpich/ticket/2205 My question : 1> What happens if we don't call MPI_Finalize and call exit(0)? 2> Is there anyway I can forcefully complete MPI_disconnect from server side ? Thanks, Hirak Ps : The reason of using sock is : bug in nemesis : http://trac.mpich.org/projects/mpich/ticket/1103 and http://trac.mpich.org/projects/mpich/ticket/79 Wesley Bland wbland at anl.gov <mailto:discuss%40mpich.org?Subject=Re%3A%20%5Bmpich-discuss%5D%20MPI_Finalize%20hangs%20in%20dynamic%20connection%20in%0A%09case%20of%20failed%20process&In-Reply-To=%3CF4B1C63E-062B-4036-9DE5-A4C93096F32C%40anl.gov%3E> Thu Feb 26 10:13:42 CST 2015 * Previous message: [mpich-discuss] MPI_Finalize hangs in dynamic connection in case of failed process<http://lists.mpich.org/pipermail/discuss/2015-February/003725.html> * Next message: [mpich-discuss] query ABI version<http://lists.mpich.org/pipermail/discuss/2015-February/003726.html> * Messages sorted by: [ date ]<http://lists.mpich.org/pipermail/discuss/2015-February/date.html#3740> [ thread ]<http://lists.mpich.org/pipermail/discuss/2015-February/thread.html#3740> [ subject ]<http://lists.mpich.org/pipermail/discuss/2015-February/subject.html#3740> [ author ]<http://lists.mpich.org/pipermail/discuss/2015-February/author.html#3740> ________________________________ First, I believe the sock device is untested with most of the MPICH fault tolerance features, so YMMV here. Is there a reason that you aren't calling MPI_Disconnect for the failed process? Did you try it an something bad happened? That seems like the most straightforward way of doing things. Otherwise, this sounds like a known issue that we're seeing from time to time with MPI_Finalize and the FT work. It's something I'm trying to figure out now. If you can reduce your code down to the minimum and send it to me, I can use it as a test case to try to fix the problem. Thanks, Wesley > On Feb 19, 2015, at 5:15 AM, Roy, Hirak <Hirak_Roy at mentor.com<https://lists.mpich.org/mailman/listinfo/discuss>> wrote: > > Hi All, > > I am using MPICH with sock connection. > I also setup processes using dynamic connection method (MPI_Comm_connect/MPI_Comm_accept). It's a master-slave architecture where master accepts the connections from slaves. > > Now if one of the process dies (or get killed), I can still recover from this (without using checkpoint/restore method). > For the particular process in master, I do not call MPI_disconnect (it hangs and does not complete). > As a result, my MPI_Finalize in master hangs and does not complete. > Do you have a workaround to forcefully complete MPI_Finalize or MPI_disconnect? > I tried MPI_Comm_free on the failed connection. However, it does not solve the hang in finalize. > > Thanks, > Hirak > _______________________________________________ _______________________________________________ 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] MPI scatter algorithms for BlueGene/P
by Jeff Hammond 02 Mar '15

02 Mar '15

Those operations were never specially optimized on Blue Gene. The operations that were optimized heavily on Blue Gene were/are Barrier, Bcast, Allreduce, and Alltoall(v). Most of the other collective optimizations were derivatives of that. For example, one can implement Allgather as a series of Bcasts. And Reduce was optimized as a side effect of Allreduce; in some cases, Allreduce was faster than Reduce, which is counter-intuitive. One optimization that Blue Gene had for Gather was to use Reduce with BOR (on BG/P) and SUM (on BG/Q, for float types at least). This turned Gather of count=1 into Reduce of count=nproc (size of communicator), but it was very faster for short messages. There aren't many good optimizations for Scatter. MPICH has most of the generic ones, as you might expect. It is possible that Scatter as Alltoallv with only one non-zero in the count vector was faster than MPICH, but this would surprise me. In any case, all MPI collectives on BG benefit from very good network-to-processor balance, minimal rendezvous (connectionless HW and SW), trivial virtual-to-physical translation, good bisection bandwidth of _electrically isolated_ torus networks, etc. If you have a more specific question, I might have a better answer. Best, Jeff On Sun, Mar 1, 2015 at 4:36 PM, Khalid Hasanov <xalid.h(a)gmail.com> wrote: > Hello, > > First of all, I am not sure if this group is the right place for this > question. If not I apologize for asking unrelated question. > > I read two papers about optimizing MPI collective communications on BG/L and > BG/P. > (Optimization of MPI Collective Communication on BlueGene/L Systems > and MPI Collective Communications on The Blue Gene/P Supercomputer: > Algorithms and Optimizations respectively). However, these two papers do not > mention anything about MPI scatter and gather operations, I wonder if these > two collective operations have been optimized for BlueGene or they use > exactly the same algorithms from MPICH. Any reference appreciated. > > > Best regards, > Khalid > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/ _______________________________________________ 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] MPICH V3.0.4 - Unable to compile simple code
by Jeff Hammond 02 Mar '15

02 Mar '15

For example, it helps to know what compiler you are using. Since MPICH contains absolutely no C++ outside of the now-deleted C++ bindings, that doesn't matter. I can only assume you are using the MPI C bindings. In the rare instance of Linux-related issues, the result of "uname -a" contains the important info. I've never seen an MPICH bug that was specific to the Linux distro... Just attach your config.log and that will provide the information people need. Jeff On Mon, Mar 2, 2015 at 1:54 PM, Kush Patel <kushpatel000(a)gmail.com> wrote: > I'm not sure exactly what information you need, but here is what I know. > > Environment: Debian > MPICH version: 3.0.4 > Language: C++ > > In regards to that link, is the process just replacing the indicated source > files and reinstalling MPICH? Also it seems that issue was for Fortran90, > would the same procedure described work for C++? > > Thank you. > > On Mon, Mar 2, 2015 at 3:34 PM, Jeff Hammond <jeff.science(a)gmail.com> wrote: >> >> You're going to need to provide more information for someone to help you. >> >> I see this issue has been encountered in the past with PGI: >> http://trac.mpich.org/projects/mpich/ticket/1073. >> >> Jeff >> >> On Mon, Mar 2, 2015 at 1:31 PM, Kush Patel <kushpatel000(a)gmail.com> wrote: >> > Hello, >> > >> > I'm having trouble getting a simple "Hello World" code to compile. The >> > following errors occur when I try to call MPI_INIT. >> > >> > /usr/bin/ld: Warning: alignment 16 of symbol `_Cmpifcmb3' in >> > /usr/local/lib/libmpich.a(setbot.o) is smaller than 32 in >> > /usr/local/lib/libmpich.a(setbotf.o) >> > /usr/bin/ld: Warning: alignment 16 of symbol `_Cmpifcmb5' in >> > /usr/local/lib/libmpich.a(setbot.o) is smaller than 32 in >> > /usr/local/lib/libmpich.a(setbotf.o) >> > /usr/bin/ld: Warning: alignment 16 of symbol `_Cmpifcmb7' in >> > /usr/local/lib/libmpich.a(setbot.o) is smaller than 32 in >> > /usr/local/lib/libmpich.a(setbotf.o) >> > /usr/bin/ld: Warning: alignment 16 of symbol `_Cmpifcmb9' in >> > /usr/local/lib/libmpich.a(setbot.o) is smaller than 32 in >> > /usr/local/lib/libmpich.a(setbotf.o) >> > >> > I have been unable to find a solution to this. >> > >> > Thank you, >> > >> > -- >> > Kush Patel >> > University of Houston >> > >> > _______________________________________________ >> > discuss mailing list discuss(a)mpich.org >> > To manage subscription options or unsubscribe: >> > https://lists.mpich.org/mailman/listinfo/discuss >> >> >> >> -- >> Jeff Hammond >> jeff.science(a)gmail.com >> http://jeffhammond.github.io/ >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > > > > > -- > Kush Patel > Doctoral Student - Bittner Group > University of Houston > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/ _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

MPE running problem
by Mikhael C 02 Mar '15

02 Mar '15

Hello, I'm having some trouble trying to generate log files with MPE. I've installed MPICH with the following command: ./configure --prefix=/usr/local/mpich 2>&1 | tee c.txt And then the MPE: ./configure --with-mpich=--prefix=/usr/local/mpich \ --with-mpichdev=/usr/local/mpich/lib \ --prefix=/usr/local/mpich \ --with-mpicc=/usr/local/mpich/bin/mpicc \ --disable-f77 MPICH goes fine, but when i try to compile with mpecc i get this message: $mpecc -mpilog -o monte_carlo_pi_2.o monte_carlo_pi_2.c /usr/bin/ld: library not found for -llmpe collect2: error: ld returned 1 exit status The equivalent expression gives the same error: $ mpicc -o monte_carlo_pi_2.o monte_carlo_pi_2.c -L/usr/local/mpich/lib -llmpe -lmpe Can someone help me? Thank You, Mikhael Chrum UTFPR - Toledo, Brasil _______________________________________________ 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] MPICH V3.0.4 - Unable to compile simple code
by Kush Patel 02 Mar '15

02 Mar '15

I'm not sure exactly what information you need, but here is what I know. Environment: Debian MPICH version: 3.0.4 Language: C++ In regards to that link, is the process just replacing the indicated source files and reinstalling MPICH? Also it seems that issue was for Fortran90, would the same procedure described work for C++? Thank you. On Mon, Mar 2, 2015 at 3:34 PM, Jeff Hammond <jeff.science(a)gmail.com> wrote: > You're going to need to provide more information for someone to help you. > > I see this issue has been encountered in the past with PGI: > http://trac.mpich.org/projects/mpich/ticket/1073. > > Jeff > > On Mon, Mar 2, 2015 at 1:31 PM, Kush Patel <kushpatel000(a)gmail.com> wrote: > > Hello, > > > > I'm having trouble getting a simple "Hello World" code to compile. The > > following errors occur when I try to call MPI_INIT. > > > > /usr/bin/ld: Warning: alignment 16 of symbol `_Cmpifcmb3' in > > /usr/local/lib/libmpich.a(setbot.o) is smaller than 32 in > > /usr/local/lib/libmpich.a(setbotf.o) > > /usr/bin/ld: Warning: alignment 16 of symbol `_Cmpifcmb5' in > > /usr/local/lib/libmpich.a(setbot.o) is smaller than 32 in > > /usr/local/lib/libmpich.a(setbotf.o) > > /usr/bin/ld: Warning: alignment 16 of symbol `_Cmpifcmb7' in > > /usr/local/lib/libmpich.a(setbot.o) is smaller than 32 in > > /usr/local/lib/libmpich.a(setbotf.o) > > /usr/bin/ld: Warning: alignment 16 of symbol `_Cmpifcmb9' in > > /usr/local/lib/libmpich.a(setbot.o) is smaller than 32 in > > /usr/local/lib/libmpich.a(setbotf.o) > > > > I have been unable to find a solution to this. > > > > Thank you, > > > > -- > > Kush Patel > > University of Houston > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > jeff.science(a)gmail.com > http://jeffhammond.github.io/ > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > -- Kush Patel Doctoral Student - Bittner Group University of Houston _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0