From halder_raju at yahoo.com Wed Jun 1 12:54:32 2005 From: halder_raju at yahoo.com (Raju Halder) Date: Wed, 1 Jun 2005 17:54:32 +0100 (BST) Subject: [Beowulf] Call For Paper Message-ID: <20050601165432.89180.qmail@web90009.mail.scd.yahoo.com> Sir, I am Raju Halder, a student of Calcutta University.I am very interested to know about CLUSTER COMPUTING.Also it is my seminar topics as a part of B. Tech. Comp. Sc. & Engg degree. I will be greatful if you help me by sending a tutorial about that.My email ID is- halder_raju at yahoo.com. Yours faithfully Raju Halder --------------------------------- Too much spam in your inbox? Yahoo! Mail gives you the best spam protection for FREE! http://in.mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From rgb at phy.duke.edu Fri Jun 3 14:35:34 2005 From: rgb at phy.duke.edu (Robert G. Brown) Date: Fri, 3 Jun 2005 14:35:34 -0400 (EDT) Subject: [Beowulf] Call For Paper In-Reply-To: <20050601165432.89180.qmail@web90009.mail.scd.yahoo.com> References: <20050601165432.89180.qmail@web90009.mail.scd.yahoo.com> Message-ID: On Wed, 1 Jun 2005, Raju Halder wrote: > Sir, > I am Raju Halder, a student of Calcutta University.I am very interested to know about CLUSTER COMPUTING.Also it is my seminar topics as a part of B. Tech. Comp. Sc. & Engg degree. > I will be greatful if you help me by sending a tutorial about that.My email ID is- Well, I cannot send you "a tutorial", but if you look on my website (URL below) and look around under "Beowulf", or look on http://www.phy.duke.edu/brahma you will find all sorts of resources -- tutorials, online books, and links to lots of things. Hope it helps. rgb > halder_raju at yahoo.com. > Yours faithfully > Raju Halder > > > > > --------------------------------- > Too much spam in your inbox? Yahoo! Mail gives you the best spam protection for FREE! > http://in.mail.yahoo.com -- Robert G. Brown http://www.phy.duke.edu/~rgb/ Duke University Dept. of Physics, Box 90305 Durham, N.C. 27708-0305 Phone: 1-919-660-2567 Fax: 919-660-2525 email:rgb at phy.duke.edu _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From maillists at gauckler.ch Wed Jun 1 16:33:39 2005 From: maillists at gauckler.ch (Michael Gauckler) Date: Wed, 01 Jun 2005 22:33:39 +0200 Subject: [Beowulf] MPICH2 + PVFS2 + Help needed urgently. Message-ID: <1117658019.15402.1.camel@localhost.localdomain> Dear Lists, I am having problems with the performance of MPICH2 and PVFS2. The program attached below should write 136MB junks of data to a 2.7GB file on a pvfs2 mount. Unfortunately the performance is so poor that my program never finishes. PVFS2 performance seems not great but acceptable for 136 MB junks to finish soon (122MB/s, see below). If someone could run a test on his machine and give me estimation of the runtime or hints where the problem might be I would be more than happy! I need to locate the problem: Code, MPICH2, ROMIO, PVFS2. Sincereley yours, Michael ___ System configuration 40 Dual Xeon 3.0 GHz, all acting as PVFS2 data servers. GigE Ethernet. Software RAID on 2 SCSI disks. Debian Sarge: Linux 2.6.8-2-686-smp #1 SMP Mon Jan 24 02:32:52 EST 2005 i686 GNU/Linux ___ Performance of PVFS2: mpdrun -np 2 ./mpi-io-test # Using mpi-io calls. nr_procs = 2, nr_iter = 1, blk_sz = 16777216 # total_size = 33554432 # Write: min_t = 0.045768, max_t = 0.274489, mean_t = 0.160128, var_t = 0.026157 # Read: min_t = 0.023897, max_t = 0.038090, mean_t = 0.030993, var_t = 0.000101 Write bandwidth = 122.243300 Mbytes/sec Read bandwidth = 880.925184 Mbytes/sec ___ Command line to run programm given below: mpdrun -1 -np 2 ./mpicube ___ Programm "mpicube.cpp": #include "mpi.h" #include #include #include #include #include char filename[] = "pvfs2:/mnt/pvfs2/mpicube_testfile.dat"; // the following lines might not be needed if not linked with the boost library namespace boost { void assertion_failed(char const * expr, char const * function, char const * file, long line) { std::ostringstream ss; ss << "BOOST_ASSERT failed for expr " << expr << ", function " << function << " in file " << file << " at line " << line << std::endl; throw std::runtime_error(ss.str()); } } int main( int argc, char *argv[] ) { int rank; int err; int worldsize; MPI_Offset headerOffset = 0; MPI_File fh; MPI_Datatype filetype; MPI_Datatype datatype = MPI_DOUBLE; MPI_Init(&argc, &argv); MPI_Comm_rank(MPI_COMM_WORLD, &rank); MPI_Comm_size(MPI_COMM_WORLD, &worldsize); printf("Hello world from process %d of %d with filename %s\n", rank, worldsize, filename); int iterations = 10; int extent0 = 600; int extent1 = 12; int extent2 = 10; int numSamples = 5000; int numSamplesPerBlock = numSamples / worldsize / iterations; int numIterConcurrent = 1; int numFinalConcurrent = 0; int groupColor = 0; int current; int gsizes[4]; int lsizes[4]; int indices[4]; gsizes[0] = extent0; gsizes[1] = extent1; gsizes[2] = extent2; gsizes[3] = numSamples; lsizes[0] = extent0; lsizes[1] = extent1; lsizes[2] = extent2; lsizes[3] = numSamplesPerBlock; indices[0] = 0; indices[1] = 0; indices[2] = 0; MPI_Comm groupcomm = MPI_COMM_WORLD; std::cout << "opening file <" << filename << ">" << std::flush << std::endl; MPI_File_open(groupcomm, filename, MPI_MODE_RDWR | MPI_MODE_CREATE | MPI_MODE_UNIQUE_OPEN, MPI_INFO_NULL, &fh); std::cout << "opened file" << std::flush << std::endl; // number of elements of type T to be stored long long lcubesize = lsizes[0]*lsizes[1]*lsizes[2]*lsizes[3]; long long gcubesize = gsizes[0]*gsizes[1]*gsizes[2]*gsizes[3]; std::cout << "local cube size * 8 = " << (long long)lcubesize / 1024 / 1024 * 8 << " MB " << std::flush << std::endl; std::cout << "global cube size * 8 = " << (long long)gcubesize / 1024 / 1024 * 8 << " MB " << std::flush << std::endl; double *cube = new double[extent0 * extent1 * extent2 * numSamplesPerBlock]; for(int j = 0; j < extent0 * extent1 * extent2 * numSamplesPerBlock; j++) cube[j] = 3.1415; for(int i = 0; i < iterations; i++){ indices[3] = (i + rank*iterations)*numSamplesPerBlock; std::cout << "iteration = " << i << std::endl; std::cout << "indices[3] = " << indices[3] << std::endl; // create a data type to get desired view of file err = MPI_Type_create_subarray(4, gsizes, lsizes, indices, MPI_ORDER_C, MPI_DOUBLE, &filetype); if (err != MPI_SUCCESS) std::cerr << "could not create subarray" << std::endl; err = MPI_Type_commit(&filetype); if (err != MPI_SUCCESS) std::cerr << "could not commit datatype" << std::endl; std::cout << "writeSubCube: setting view" << std::endl; // store the view into file err = MPI_File_set_view(fh, 0, datatype, filetype, "native", MPI_INFO_NULL); if (err != MPI_SUCCESS) std::cerr << "could not set view" << std::endl; std::cout << "allocating cube" << std::endl; std::cout << "starting write all" << std::endl; err = MPI_File_write_all(fh, &cube[0], lcubesize, datatype, MPI_STATUS_IGNORE); if (err != MPI_SUCCESS) std::cerr << "could not write to file" << std::endl; std::cout << "done write all" << std::endl; err = MPI_Type_free(&filetype); if (err != MPI_SUCCESS) std::cerr << "could not free datatype" << std::endl; } MPI_File_close(&fh); std::cout << "closed file" << std::flush << std::endl; MPI_Finalize(); return 0; } _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From mwill at penguincomputing.com Mon Jun 6 02:43:06 2005 From: mwill at penguincomputing.com (Michael Will) Date: Sun, 05 Jun 2005 23:43:06 -0700 Subject: [Beowulf] MPICH2 + PVFS2 + Help needed urgently. In-Reply-To: <1117658019.15402.1.camel@localhost.localdomain> References: <1117658019.15402.1.camel@localhost.localdomain> Message-ID: <42A3F07A.4080806@penguincomputing.com>  I have not had a chance to test anything yet, but my first guess is that you are not having realistic benchmark numbers to begin with. You say you see 122MB/s write, 880MB/s read? That for sure is not across gigabit ethernet unless you have eight NICs on the reading node ;-) If you were using infiniband this might be realistic, but for gigabit ethernet it is not. Viel Glueck, Michael Michael Gauckler wrote: >Dear Lists, > >I am having problems with the performance of MPICH2 and PVFS2. > >The program attached below should write 136MB junks of data to a >2.7GB file on a pvfs2 mount. > >Unfortunately the performance is so poor that my program never >finishes. PVFS2 performance seems not great but acceptable for >136 MB junks to finish soon (122MB/s, see below). > >If someone could run a test on his machine and give me estimation of >the runtime or hints where the problem might be I would be more than >happy! I need to locate the problem: Code, MPICH2, ROMIO, PVFS2. > >Sincereley yours, >Michael > > -- Michael Will Penguin Computing Corp. Sales Engineer 415-954-2887 415-954-2899 fx mwill at penguincomputing.com Visit us at the following Linux Shows! Bio-IT World Conference and Expo '05 Hynes Convention Center, Boston, MA May 17th-19th, 2005 Booth 201 _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From alangrimes at starpower.net Thu Jun 30 12:30:34 2005 From: alangrimes at starpower.net (Alan Grimes) Date: Thu, 30 Jun 2005 11:30:34 -0500 Subject: [Beowulf] more news on the Cell In-Reply-To: <20050630141244.GL25947@leitl.org> References: <20050630141244.GL25947@leitl.org> Message-ID: <42C41E2A.4060707@starpower.net> Eugen Leitl wrote: >http://www-128.ibm.com/developerworks/power/library/pa-cell/?ca=dgr-pow03SpufsCell > >also, see http://www.research.scea.com/research/html/CellGDC05/index.html > > I'm sick of reading this type of stuff... I want to see actual mobos/cards that use this chip! Everything that could possibly be said about Cell is moot without available hardware! _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From federico.ceccarelli at techcom.it Thu Jun 30 08:34:31 2005 From: federico.ceccarelli at techcom.it (Federico Ceccarelli) Date: Thu, 30 Jun 2005 14:34:31 +0200 Subject: [Beowulf] WRF model on linux cluster: Mpi problem In-Reply-To: <3.0.32.20050630124421.0111d118@pop3.xs4all.nl> References: <3.0.32.20050630124421.0111d118@pop3.xs4all.nl> Message-ID: <1120134871.5110.29.camel@localhost.localdomain> Thanks for you answer Vincent, my network cards are Intel Pro 1000, Gigabit. Yes I did a 72h (real time) simulations that lasted 20h on 4 cpus...same behaviour... I'm thinking about a bandwith problem... ....maybe due to hardware failure of some network card, or switch (3com -Baseline switch 2824). Or the pci-raisers for the network card (I have a 2 unit rack so that I cannot mount network cards directly on the pci slot)... Did you experience problem with pci-raisers? Can you suggest me a bandwidth benchmark? thanks again... federico Il giorno gio, 30-06-2005 alle 12:44 +0200, Vincent Diepeveen ha scritto: > Hello Federico, > > Hope you can find contacts to colleges. > > A few questions. > a) what kind of interconnects does the cluster have (networkcards and > which type?) > b) if you run a simulation that eats a few hours instead of a few seconds, > do you get the same speed outcome difference? > > I see the program is pretty big for open source calculating software, about > 1.9MB fortran code, so bit time consuming to figure out for someone who > isn't a non-meteorological expert. > > E:\wrf>dir *.f* /s /p > .. > Total Files Listed: > 141 File(s) 1,972,938 bytes > > Best regards, > Vincent > > At 06:56 PM 6/29/2005 +0200, federico.ceccarelli wrote: > > > >Hi! > > > >I would like to get in touch with people running numerical meteorological > >models on a linux cluster (16cpu) , distributed memory (1Gb every node), > >diskless nodes, Gigabit lan, mpich and openmosix. > > > >I'm tring to run WRF model but the mpi version parallelized on 4, 8, or 16 > >nodes runs slower than the single node one! It runs correctly but so slow... > > > >When I run wrf.exe on a single processor the cpu time for every timestep is > >about 10s for my configuration. > > > >When I switch to np=4, 8 or 16 the cpu time for a single step sometimes its > >faster (as It should always be, for example 3sec for 4 cpu ) but often it is > >slower and slower (60sec and more!). The overall time of the simulation is > >bigger than for the single node run... > > > >anyone have experienced the same problem? > > > >thanks in advance to everybody... > > > >federico > > > > > > > >Dr. Federico Ceccarelli (PhD) > >----------------------------- > > TechCom snc > >Via di Sottoripa 1-18 > >16124 Genova - Italia > >Tel: +39 010 860 5664 > >Fax: +39 010 860 5691 > >http://www.techcom.it > > > >_______________________________________________ > >Beowulf mailing list, Beowulf at beowulf.org > >To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf > > > > > _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From romieu at fr.zoreil.com Thu Jun 30 11:32:52 2005 From: romieu at fr.zoreil.com (Francois Romieu) Date: Thu, 30 Jun 2005 17:32:52 +0200 Subject: [Beowulf] Realtek 8169 performance problem In-Reply-To: <002101c57819$e5402930$7a13a296@labplan22> References: <002101c57819$e5402930$7a13a296@labplan22> Message-ID: <20050630153252.GA7706@se1.cogenit.fr> Daniel Dotta : [...] > So I would like to know if anyone had a similar performance problem with > this network interface? Would be a diferent configuration for the network > interface or when I operate in Gigabit mode? Someone could suggest a good > gigabit network interface for we test and compare the performance? You can send a PR to the r8169 maintainer including: - a test scenario - some figures - the .config - dmesg from boot - lspci -vx output - lsmod output - ifconfig output - the content of /proc/interrupts (a tarball or an url are both fine with me) If you do so, please keep some public mailing list in the Cc: (beowulf or netdev at vger.kernel.org for instance). -- Ueimor _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From jdvw at tticluster.com Thu Jun 30 15:57:53 2005 From: jdvw at tticluster.com (John Van Workum) Date: Thu, 30 Jun 2005 15:57:53 -0400 (EDT) Subject: [Beowulf] WRF model on linux cluster: Mpi problem In-Reply-To: <200506301900.j5UJ0DFo010318@bluewest.scyld.com> References: <200506301900.j5UJ0DFo010318@bluewest.scyld.com> Message-ID: <58863.160.142.1.16.1120161473.squirrel@160.142.1.16> Federico, It could be simply a problem with the data set your are trying to process and not the cluster. WRF is a fairly well supported and written application. If you run faster on a single CPU, then I suspect that particular data set is not "easily" parallizable to run on 16. You may want to try other test cases, one that you know will run well with 16 CPUs. Once you find it, I bet it will crawl on 1 or 2 CPUs. John Van Workum TTI > > At 06:56 PM 6/29/2005 +0200, federico.ceccarelli wrote: >> >>Hi! >> >>I would like to get in touch with people running numerical meteorological >>models on a linux cluster (16cpu) , distributed memory (1Gb every node), >>diskless nodes, Gigabit lan, mpich and openmosix. >> >>I'm tring to run WRF model but the mpi version parallelized on 4, 8, or 16 >>nodes runs slower than the single node one! It runs correctly but so slow... >> >>When I run wrf.exe on a single processor the cpu time for every timestep is >>about 10s for my configuration. >> >>When I switch to np=4, 8 or 16 the cpu time for a single step sometimes its >>faster (as It should always be, for example 3sec for 4 cpu ) but often it is >>slower and slower (60sec and more!). The overall time of the simulation is >>bigger than for the single node run... >> >>anyone have experienced the same problem? >> >>thanks in advance to everybody... >> >>federico >> _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From halder_raju at yahoo.com Wed Jun 1 12:54:32 2005 From: halder_raju at yahoo.com (Raju Halder) Date: Wed, 1 Jun 2005 17:54:32 +0100 (BST) Subject: [Beowulf] Call For Paper Message-ID: <20050601165432.89180.qmail@web90009.mail.scd.yahoo.com> Sir, I am Raju Halder, a student of Calcutta University.I am very interested to know about CLUSTER COMPUTING.Also it is my seminar topics as a part of B. Tech. Comp. Sc. & Engg degree. I will be greatful if you help me by sending a tutorial about that.My email ID is- halder_raju at yahoo.com. Yours faithfully Raju Halder --------------------------------- Too much spam in your inbox? Yahoo! Mail gives you the best spam protection for FREE! http://in.mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From rgb at phy.duke.edu Fri Jun 3 14:35:34 2005 From: rgb at phy.duke.edu (Robert G. Brown) Date: Fri, 3 Jun 2005 14:35:34 -0400 (EDT) Subject: [Beowulf] Call For Paper In-Reply-To: <20050601165432.89180.qmail@web90009.mail.scd.yahoo.com> References: <20050601165432.89180.qmail@web90009.mail.scd.yahoo.com> Message-ID: On Wed, 1 Jun 2005, Raju Halder wrote: > Sir, > I am Raju Halder, a student of Calcutta University.I am very interested to know about CLUSTER COMPUTING.Also it is my seminar topics as a part of B. Tech. Comp. Sc. & Engg degree. > I will be greatful if you help me by sending a tutorial about that.My email ID is- Well, I cannot send you "a tutorial", but if you look on my website (URL below) and look around under "Beowulf", or look on http://www.phy.duke.edu/brahma you will find all sorts of resources -- tutorials, online books, and links to lots of things. Hope it helps. rgb > halder_raju at yahoo.com. > Yours faithfully > Raju Halder > > > > > --------------------------------- > Too much spam in your inbox? Yahoo! Mail gives you the best spam protection for FREE! > http://in.mail.yahoo.com -- Robert G. Brown http://www.phy.duke.edu/~rgb/ Duke University Dept. of Physics, Box 90305 Durham, N.C. 27708-0305 Phone: 1-919-660-2567 Fax: 919-660-2525 email:rgb at phy.duke.edu _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From maillists at gauckler.ch Wed Jun 1 16:33:39 2005 From: maillists at gauckler.ch (Michael Gauckler) Date: Wed, 01 Jun 2005 22:33:39 +0200 Subject: [Beowulf] MPICH2 + PVFS2 + Help needed urgently. Message-ID: <1117658019.15402.1.camel@localhost.localdomain> Dear Lists, I am having problems with the performance of MPICH2 and PVFS2. The program attached below should write 136MB junks of data to a 2.7GB file on a pvfs2 mount. Unfortunately the performance is so poor that my program never finishes. PVFS2 performance seems not great but acceptable for 136 MB junks to finish soon (122MB/s, see below). If someone could run a test on his machine and give me estimation of the runtime or hints where the problem might be I would be more than happy! I need to locate the problem: Code, MPICH2, ROMIO, PVFS2. Sincereley yours, Michael ___ System configuration 40 Dual Xeon 3.0 GHz, all acting as PVFS2 data servers. GigE Ethernet. Software RAID on 2 SCSI disks. Debian Sarge: Linux 2.6.8-2-686-smp #1 SMP Mon Jan 24 02:32:52 EST 2005 i686 GNU/Linux ___ Performance of PVFS2: mpdrun -np 2 ./mpi-io-test # Using mpi-io calls. nr_procs = 2, nr_iter = 1, blk_sz = 16777216 # total_size = 33554432 # Write: min_t = 0.045768, max_t = 0.274489, mean_t = 0.160128, var_t = 0.026157 # Read: min_t = 0.023897, max_t = 0.038090, mean_t = 0.030993, var_t = 0.000101 Write bandwidth = 122.243300 Mbytes/sec Read bandwidth = 880.925184 Mbytes/sec ___ Command line to run programm given below: mpdrun -1 -np 2 ./mpicube ___ Programm "mpicube.cpp": #include "mpi.h" #include #include #include #include #include char filename[] = "pvfs2:/mnt/pvfs2/mpicube_testfile.dat"; // the following lines might not be needed if not linked with the boost library namespace boost { void assertion_failed(char const * expr, char const * function, char const * file, long line) { std::ostringstream ss; ss << "BOOST_ASSERT failed for expr " << expr << ", function " << function << " in file " << file << " at line " << line << std::endl; throw std::runtime_error(ss.str()); } } int main( int argc, char *argv[] ) { int rank; int err; int worldsize; MPI_Offset headerOffset = 0; MPI_File fh; MPI_Datatype filetype; MPI_Datatype datatype = MPI_DOUBLE; MPI_Init(&argc, &argv); MPI_Comm_rank(MPI_COMM_WORLD, &rank); MPI_Comm_size(MPI_COMM_WORLD, &worldsize); printf("Hello world from process %d of %d with filename %s\n", rank, worldsize, filename); int iterations = 10; int extent0 = 600; int extent1 = 12; int extent2 = 10; int numSamples = 5000; int numSamplesPerBlock = numSamples / worldsize / iterations; int numIterConcurrent = 1; int numFinalConcurrent = 0; int groupColor = 0; int current; int gsizes[4]; int lsizes[4]; int indices[4]; gsizes[0] = extent0; gsizes[1] = extent1; gsizes[2] = extent2; gsizes[3] = numSamples; lsizes[0] = extent0; lsizes[1] = extent1; lsizes[2] = extent2; lsizes[3] = numSamplesPerBlock; indices[0] = 0; indices[1] = 0; indices[2] = 0; MPI_Comm groupcomm = MPI_COMM_WORLD; std::cout << "opening file <" << filename << ">" << std::flush << std::endl; MPI_File_open(groupcomm, filename, MPI_MODE_RDWR | MPI_MODE_CREATE | MPI_MODE_UNIQUE_OPEN, MPI_INFO_NULL, &fh); std::cout << "opened file" << std::flush << std::endl; // number of elements of type T to be stored long long lcubesize = lsizes[0]*lsizes[1]*lsizes[2]*lsizes[3]; long long gcubesize = gsizes[0]*gsizes[1]*gsizes[2]*gsizes[3]; std::cout << "local cube size * 8 = " << (long long)lcubesize / 1024 / 1024 * 8 << " MB " << std::flush << std::endl; std::cout << "global cube size * 8 = " << (long long)gcubesize / 1024 / 1024 * 8 << " MB " << std::flush << std::endl; double *cube = new double[extent0 * extent1 * extent2 * numSamplesPerBlock]; for(int j = 0; j < extent0 * extent1 * extent2 * numSamplesPerBlock; j++) cube[j] = 3.1415; for(int i = 0; i < iterations; i++){ indices[3] = (i + rank*iterations)*numSamplesPerBlock; std::cout << "iteration = " << i << std::endl; std::cout << "indices[3] = " << indices[3] << std::endl; // create a data type to get desired view of file err = MPI_Type_create_subarray(4, gsizes, lsizes, indices, MPI_ORDER_C, MPI_DOUBLE, &filetype); if (err != MPI_SUCCESS) std::cerr << "could not create subarray" << std::endl; err = MPI_Type_commit(&filetype); if (err != MPI_SUCCESS) std::cerr << "could not commit datatype" << std::endl; std::cout << "writeSubCube: setting view" << std::endl; // store the view into file err = MPI_File_set_view(fh, 0, datatype, filetype, "native", MPI_INFO_NULL); if (err != MPI_SUCCESS) std::cerr << "could not set view" << std::endl; std::cout << "allocating cube" << std::endl; std::cout << "starting write all" << std::endl; err = MPI_File_write_all(fh, &cube[0], lcubesize, datatype, MPI_STATUS_IGNORE); if (err != MPI_SUCCESS) std::cerr << "could not write to file" << std::endl; std::cout << "done write all" << std::endl; err = MPI_Type_free(&filetype); if (err != MPI_SUCCESS) std::cerr << "could not free datatype" << std::endl; } MPI_File_close(&fh); std::cout << "closed file" << std::flush << std::endl; MPI_Finalize(); return 0; } _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From mwill at penguincomputing.com Mon Jun 6 02:43:06 2005 From: mwill at penguincomputing.com (Michael Will) Date: Sun, 05 Jun 2005 23:43:06 -0700 Subject: [Beowulf] MPICH2 + PVFS2 + Help needed urgently. In-Reply-To: <1117658019.15402.1.camel@localhost.localdomain> References: <1117658019.15402.1.camel@localhost.localdomain> Message-ID: <42A3F07A.4080806@penguincomputing.com>  I have not had a chance to test anything yet, but my first guess is that you are not having realistic benchmark numbers to begin with. You say you see 122MB/s write, 880MB/s read? That for sure is not across gigabit ethernet unless you have eight NICs on the reading node ;-) If you were using infiniband this might be realistic, but for gigabit ethernet it is not. Viel Glueck, Michael Michael Gauckler wrote: >Dear Lists, > >I am having problems with the performance of MPICH2 and PVFS2. > >The program attached below should write 136MB junks of data to a >2.7GB file on a pvfs2 mount. > >Unfortunately the performance is so poor that my program never >finishes. PVFS2 performance seems not great but acceptable for >136 MB junks to finish soon (122MB/s, see below). > >If someone could run a test on his machine and give me estimation of >the runtime or hints where the problem might be I would be more than >happy! I need to locate the problem: Code, MPICH2, ROMIO, PVFS2. > >Sincereley yours, >Michael > > -- Michael Will Penguin Computing Corp. Sales Engineer 415-954-2887 415-954-2899 fx mwill at penguincomputing.com Visit us at the following Linux Shows! Bio-IT World Conference and Expo '05 Hynes Convention Center, Boston, MA May 17th-19th, 2005 Booth 201 _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From alangrimes at starpower.net Thu Jun 30 12:30:34 2005 From: alangrimes at starpower.net (Alan Grimes) Date: Thu, 30 Jun 2005 11:30:34 -0500 Subject: [Beowulf] more news on the Cell In-Reply-To: <20050630141244.GL25947@leitl.org> References: <20050630141244.GL25947@leitl.org> Message-ID: <42C41E2A.4060707@starpower.net> Eugen Leitl wrote: >http://www-128.ibm.com/developerworks/power/library/pa-cell/?ca=dgr-pow03SpufsCell > >also, see http://www.research.scea.com/research/html/CellGDC05/index.html > > I'm sick of reading this type of stuff... I want to see actual mobos/cards that use this chip! Everything that could possibly be said about Cell is moot without available hardware! _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From federico.ceccarelli at techcom.it Thu Jun 30 08:34:31 2005 From: federico.ceccarelli at techcom.it (Federico Ceccarelli) Date: Thu, 30 Jun 2005 14:34:31 +0200 Subject: [Beowulf] WRF model on linux cluster: Mpi problem In-Reply-To: <3.0.32.20050630124421.0111d118@pop3.xs4all.nl> References: <3.0.32.20050630124421.0111d118@pop3.xs4all.nl> Message-ID: <1120134871.5110.29.camel@localhost.localdomain> Thanks for you answer Vincent, my network cards are Intel Pro 1000, Gigabit. Yes I did a 72h (real time) simulations that lasted 20h on 4 cpus...same behaviour... I'm thinking about a bandwith problem... ....maybe due to hardware failure of some network card, or switch (3com -Baseline switch 2824). Or the pci-raisers for the network card (I have a 2 unit rack so that I cannot mount network cards directly on the pci slot)... Did you experience problem with pci-raisers? Can you suggest me a bandwidth benchmark? thanks again... federico Il giorno gio, 30-06-2005 alle 12:44 +0200, Vincent Diepeveen ha scritto: > Hello Federico, > > Hope you can find contacts to colleges. > > A few questions. > a) what kind of interconnects does the cluster have (networkcards and > which type?) > b) if you run a simulation that eats a few hours instead of a few seconds, > do you get the same speed outcome difference? > > I see the program is pretty big for open source calculating software, about > 1.9MB fortran code, so bit time consuming to figure out for someone who > isn't a non-meteorological expert. > > E:\wrf>dir *.f* /s /p > .. > Total Files Listed: > 141 File(s) 1,972,938 bytes > > Best regards, > Vincent > > At 06:56 PM 6/29/2005 +0200, federico.ceccarelli wrote: > > > >Hi! > > > >I would like to get in touch with people running numerical meteorological > >models on a linux cluster (16cpu) , distributed memory (1Gb every node), > >diskless nodes, Gigabit lan, mpich and openmosix. > > > >I'm tring to run WRF model but the mpi version parallelized on 4, 8, or 16 > >nodes runs slower than the single node one! It runs correctly but so slow... > > > >When I run wrf.exe on a single processor the cpu time for every timestep is > >about 10s for my configuration. > > > >When I switch to np=4, 8 or 16 the cpu time for a single step sometimes its > >faster (as It should always be, for example 3sec for 4 cpu ) but often it is > >slower and slower (60sec and more!). The overall time of the simulation is > >bigger than for the single node run... > > > >anyone have experienced the same problem? > > > >thanks in advance to everybody... > > > >federico > > > > > > > >Dr. Federico Ceccarelli (PhD) > >----------------------------- > > TechCom snc > >Via di Sottoripa 1-18 > >16124 Genova - Italia > >Tel: +39 010 860 5664 > >Fax: +39 010 860 5691 > >http://www.techcom.it > > > >_______________________________________________ > >Beowulf mailing list, Beowulf at beowulf.org > >To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf > > > > > _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From romieu at fr.zoreil.com Thu Jun 30 11:32:52 2005 From: romieu at fr.zoreil.com (Francois Romieu) Date: Thu, 30 Jun 2005 17:32:52 +0200 Subject: [Beowulf] Realtek 8169 performance problem In-Reply-To: <002101c57819$e5402930$7a13a296@labplan22> References: <002101c57819$e5402930$7a13a296@labplan22> Message-ID: <20050630153252.GA7706@se1.cogenit.fr> Daniel Dotta : [...] > So I would like to know if anyone had a similar performance problem with > this network interface? Would be a diferent configuration for the network > interface or when I operate in Gigabit mode? Someone could suggest a good > gigabit network interface for we test and compare the performance? You can send a PR to the r8169 maintainer including: - a test scenario - some figures - the .config - dmesg from boot - lspci -vx output - lsmod output - ifconfig output - the content of /proc/interrupts (a tarball or an url are both fine with me) If you do so, please keep some public mailing list in the Cc: (beowulf or netdev at vger.kernel.org for instance). -- Ueimor _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From jdvw at tticluster.com Thu Jun 30 15:57:53 2005 From: jdvw at tticluster.com (John Van Workum) Date: Thu, 30 Jun 2005 15:57:53 -0400 (EDT) Subject: [Beowulf] WRF model on linux cluster: Mpi problem In-Reply-To: <200506301900.j5UJ0DFo010318@bluewest.scyld.com> References: <200506301900.j5UJ0DFo010318@bluewest.scyld.com> Message-ID: <58863.160.142.1.16.1120161473.squirrel@160.142.1.16> Federico, It could be simply a problem with the data set your are trying to process and not the cluster. WRF is a fairly well supported and written application. If you run faster on a single CPU, then I suspect that particular data set is not "easily" parallizable to run on 16. You may want to try other test cases, one that you know will run well with 16 CPUs. Once you find it, I bet it will crawl on 1 or 2 CPUs. John Van Workum TTI > > At 06:56 PM 6/29/2005 +0200, federico.ceccarelli wrote: >> >>Hi! >> >>I would like to get in touch with people running numerical meteorological >>models on a linux cluster (16cpu) , distributed memory (1Gb every node), >>diskless nodes, Gigabit lan, mpich and openmosix. >> >>I'm tring to run WRF model but the mpi version parallelized on 4, 8, or 16 >>nodes runs slower than the single node one! It runs correctly but so slow... >> >>When I run wrf.exe on a single processor the cpu time for every timestep is >>about 10s for my configuration. >> >>When I switch to np=4, 8 or 16 the cpu time for a single step sometimes its >>faster (as It should always be, for example 3sec for 4 cpu ) but often it is >>slower and slower (60sec and more!). The overall time of the simulation is >>bigger than for the single node run... >> >>anyone have experienced the same problem? >> >>thanks in advance to everybody... >> >>federico >> _______________________________________________ Beowulf mailing list, Beowulf at beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf