[gmx-users] Mismatching number of PP MPI processes and GPUs per node
Hello Sorry for posting this again. I am trying to run GROMACS 4.6 compiled with MPI and GPU acceleration (CUDA 5.0 lib) using the following SGE batch script. #!/bin/sh #$ -V #$ -S /bin/sh #$ -N test-gpus #$ -l h=xgrid-node02 #$ -pe mpi_fill_up 12 #$ -cwd source /opt/NetUsers/pgkeka/gromacs-4.6_gpu_mpi/bin/GMXRC export DYLD_LIBRARY_PATH=/Developer/NVIDIA/CUDA-5.0/lib:$DYLD_LIBRARY_PATH mpirun -np 12 mdrun_mpi -s test.tpr -deffnm test_out -nb gpu After detection of the installed GPU card 1 GPU detected on host xgrid-node02.xgrid: #0: NVIDIA Quadro 4000, compute cap.: 2.0, ECC: no, stat: compatible GROMACS issues the following error Incorrect launch configuration: mismatching number of PP MPI processes and GPUs per node. mdrun_mpi was started with 12 PP MPI processes per node, but only 1 GPU were detected. It can't be that we need to run GROMACS only on a single core so that it matches the single GPU card. Do you have any idea what has to be done? Many thanks. Dr. George Patargias Postdoctoral Researcher Biomedical Research Foundation Academy of Athens 4, Soranou Ephessiou 115 27 Athens Greece Office: +302106597568 -- gmx-users mailing listgmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
Re: [gmx-users] Mismatching number of PP MPI processes and GPUs per node
Hi, On Mar 11, 2013, at 10:50 AM, George Patargias g...@bioacademy.gr wrote: Hello Sorry for posting this again. I am trying to run GROMACS 4.6 compiled with MPI and GPU acceleration (CUDA 5.0 lib) using the following SGE batch script. #!/bin/sh #$ -V #$ -S /bin/sh #$ -N test-gpus #$ -l h=xgrid-node02 #$ -pe mpi_fill_up 12 #$ -cwd source /opt/NetUsers/pgkeka/gromacs-4.6_gpu_mpi/bin/GMXRC export DYLD_LIBRARY_PATH=/Developer/NVIDIA/CUDA-5.0/lib:$DYLD_LIBRARY_PATH mpirun -np 12 mdrun_mpi -s test.tpr -deffnm test_out -nb gpu After detection of the installed GPU card 1 GPU detected on host xgrid-node02.xgrid: #0: NVIDIA Quadro 4000, compute cap.: 2.0, ECC: no, stat: compatible GROMACS issues the following error Incorrect launch configuration: mismatching number of PP MPI processes and GPUs per node. mdrun_mpi was started with 12 PP MPI processes per node, but only 1 GPU were detected. It can't be that we need to run GROMACS only on a single core so that it matches the single GPU card. Have you compiled mdrun_mpi with OpenMP threads support? Then, if you do mpirun -np 1 mdrun_mpi … it should start one MPI process with 12 OpenMP threads, which should give you what you want. You can also manually specify the number of OpenMP threads by adding -ntomp 12 Carsten Do you have any idea what has to be done? Many thanks. Dr. George Patargias Postdoctoral Researcher Biomedical Research Foundation Academy of Athens 4, Soranou Ephessiou 115 27 Athens Greece Office: +302106597568 -- gmx-users mailing listgmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists -- Dr. Carsten Kutzner Max Planck Institute for Biophysical Chemistry Theoretical and Computational Biophysics Am Fassberg 11, 37077 Goettingen, Germany Tel. +49-551-2012313, Fax: +49-551-2012302 http://www.mpibpc.mpg.de/grubmueller/kutzner http://www.mpibpc.mpg.de/grubmueller/sppexa -- gmx-users mailing listgmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
RE: [gmx-users] mdrun WARING and crash
Hallo Justin, Thank you for your comments. Taking your suggestions, I set nrexcl=1, and comment the [pairs] section, because there is no special case of non-bonded interactions to declare, then we try to see what happens. Now we minimize it by steep, then by cg, both of the processor are very quick done, because after around 7000 steps, the energy can not go further down any more. Then we finish the mdrun, the energies output are like: Step Time Lambda 1 10.00.0 Energies (kJ/mol) LJ (SR) Coulomb (SR) PotentialKinetic En. Total Energy 1.49468e+050.0e+001.49468e+058.73890e+031.58207e+05 Temperature Pressure (bar) 4.38208e+029.93028e+04 Although this time no running error, we find the outputs are extremely wired, for example through VMD conf.gro traj.xtc we find 0 frame a homogeneous box, starting from the first step, the box becomes a lattice, which is far away from our expected the Polymer melt system should be. The force parameters are taken from literature PRL 85(5), 1128(2000), I am still very worried about the format of my input files. Could you please give me, a very beginner a help. Thanks a lot. Kind regards, Li From: gmx-users-boun...@gromacs.org [gmx-users-boun...@gromacs.org] on behalf of Justin Lemkul [jalem...@vt.edu] Sent: Thursday, February 28, 2013 3:02 PM To: Discussion list for GROMACS users Subject: Re: [gmx-users] mdrun WARING and crash On 2/28/13 6:59 AM, l@utwente.nl wrote: Hallo Justin, Thank you for you help. I have read the previous discussions on this topic, which is very helpful. The link is: http://gromacs.5086.n6.nabble.com/What-is-the-purpose-of-the-pairs-section-td5003598.html Well, there are still something I want to make sure, which might be the reason of mdrun crash of my system. ###Introduction of system## Linear Polyethylene melt: each chain contains 16 beads, each bead coarse grained 3 monomers. Number of chain in the box is 64. Force Field## ffbonded.itp [ bondtypes ] ; FENE, Ro = 1.5 sigma and kb = 30 epsilon/sigma^2 ; ij funcb0 (nm) kb (kJ/mol nm^2) CH2 CH27 0.795 393. ffnonbonded.itp [ atomtypes ] ; epsilon / kB = 443K ;name at.num mass (au) charge ptype sigma (nm)epsilon (kJ/mol) CH2 142.3 0.000 A 0.5300 3.68133 [ nonbond_params ] ; i jfunc sigma epsilon CH2 CH210.5303.68133 [ pairtypes ] ; i jfunc sigma epsilon CH2 CH21 0.533.68133 topology## [ defaults ] ; nbfunccomb-rule gen-pairs fudgeLJ fudgeQQ 1 2no 1.0 1.0 ; The force field files to be included #include ../forcefield/forcefield.itp [ moleculetype ] ; name nrexcl PE 0 [atoms] ; nrtype resnr residuatomcgnr charge 1 CH2 1PE C 1 0.0 2 CH2 1PE C 2 0.0 3 CH2 1PE C 3 0.0 4 CH2 1PE C 4 0.0 .. 15CH2 1PE C 15 0.0 16CH2 1PE C 16 0.0 [ bonds ] ; aiaj funct c0 c1 1 2 7 0.795 393. 2 3 7 0.795 393. 3 4 7 0.795 393. .. 1415 7 0.795 393. 1516 7 0.795 393. [ pairs ] ; aiaj funct c0 c1 1 2 1 0.53 3.68133 1 3 1 0.53 3.68133 1 4 1 0.53 3.68133 2 3 1 0.53 3.68133 2 4 1 0.53 3.68133 2 5 1 0.53 3.68133 .. 1415 1 0.53 3.68133 1416 1 0.53 3.68133 1516 1 0.53 3.68133 ##mdp### ;directories to include in your topology. Format include = -I/home/otterw/Install/gromacs-4.6/src/gmxlib integrator = md ; ; RUN CONTROL * ; tinit = 0; start time to run dt = 1e-3
Re: [gmx-users] mdrun WARING and crash
On Monday, March 11, 2013, wrote: Hallo Justin, Thank you for your comments. Taking your suggestions, I set nrexcl=1, and comment the [pairs] section, because there is no special case of non-bonded interactions to declare, then we try to see what happens. Now we minimize it by steep, then by cg, both of the processor are very quick done, because after around 7000 steps, the energy can not go further down any more. Then we finish the mdrun, the energies output are like: Step Time Lambda 1 10.00.0 Energies (kJ/mol) LJ (SR) Coulomb (SR) PotentialKinetic En. Total Energy 1.49468e+050.0e+001.49468e+058.73890e+031.58207e+05 Temperature Pressure (bar) 4.38208e+029.93028e+04 Although this time no running error, we find the outputs are extremely wired, for example through VMD conf.gro traj.xtc we find 0 frame a homogeneous box, starting from the first step, the box becomes a lattice, which is far away from our expected the Polymer melt system should be. The force parameters are taken from literature PRL 85(5), 1128(2000), I am still very worried about the format of my input files. Could you please give me, a very beginner a help. Please provide links to images. This is probably not a big deal as long as the simulation is actually running, since a triclinic representation of the unit cell is used. -Justin Thanks a lot. Kind regards, Li From: gmx-users-boun...@gromacs.org javascript:; [ gmx-users-boun...@gromacs.org javascript:;] on behalf of Justin Lemkul [ jalem...@vt.edu javascript:;] Sent: Thursday, February 28, 2013 3:02 PM To: Discussion list for GROMACS users Subject: Re: [gmx-users] mdrun WARING and crash On 2/28/13 6:59 AM, l@utwente.nl wrote: Hallo Justin, Thank you for you help. I have read the previous discussions on this topic, which is very helpful. The link is: http://gromacs.5086.n6.nabble.com/What-is-the-purpose-of-the-pairs-section-td5003598.html Well, there are still something I want to make sure, which might be the reason of mdrun crash of my system. ###Introduction of system## Linear Polyethylene melt: each chain contains 16 beads, each bead coarse grained 3 monomers. Number of chain in the box is 64. Force Field## ffbonded.itp [ bondtypes ] ; FENE, Ro = 1.5 sigma and kb = 30 epsilon/sigma^2 ; ij funcb0 (nm) kb (kJ/mol nm^2) CH2 CH27 0.795 393. ffnonbonded.itp [ atomtypes ] ; epsilon / kB = 443K ;name at.num mass (au) charge ptype sigma (nm) epsilon (kJ/mol) CH2 142.3 0.000 A 0.5300 3.68133 [ nonbond_params ] ; i jfunc sigma epsilon CH2 CH210.5303.68133 [ pairtypes ] ; i jfunc sigma epsilon CH2 CH21 0.533.68133 topology## [ defaults ] ; nbfunccomb-rule gen-pairs fudgeLJ fudgeQQ 1 2no 1.0 1.0 ; The force field files to be included #include ../forcefield/forcefield.itp [ moleculetype ] ; name nrexcl PE 0 [atoms] ; nrtype resnr residuatomcgnr charge 1 CH2 1PE C 1 0.0 2 CH2 1PE C 2 0.0 3 CH2 1PE C 3 0.0 4 CH2 1PE C 4 0.0 .. 15CH2 1PE C 15 0.0 16CH2 1PE C 16 0.0 [ bonds ] ; aiaj funct c0 c1 1 2 7 0.795 393. 2 3 7 0.795 393. 3 4 7 0.795 393. .. 1415 7 0.795 393. 1516 7 0.795 393. [ pairs ] ; ai -- gmx-users mailing listgmx-users@gromacs.org javascript:; http://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org javascript:;. * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists -- gmx-users mailing listgmx-users@gromacs.org javascript:; http://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to
Re: [gmx-users] only 4 zero eigvals for NMA (with 10^-6 emtol for l-bfgs)
I printed all small 50 eigenvalues, and their first 9 are listed as follows as before:* -1.8476593e-06* -1.2164109e-06* -1.4830436e-07* 1.3104790e-06* 1.1364704e+00* 1.7418209e+00* 2.2836573e+00* 3.7218206e+00* 4.4875873e+00 As you see, the first three of them are negative and close to zero, which can be considered as zero. do you know any reason of this, having only 4 (close to) zero eigenvalues ?? -- Hyuntae Message: 3 Date: Sun, 10 Mar 2013 23:10:05 +0100 From: lloyd riggs lloyd.ri...@gmx.ch Subject: Re: [gmx-users] only 4 zero eigvals for NMA (with 10^-6 emtol for l-bfgs) To: Discussion list for GROMACS users gmx-users@gromacs.org Message-ID: 20130310221005.162...@gmx.net Content-Type: text/plain; charset=utf-8 I can imagine why you would go past the first few, but does it print the zero's or just negate them from the equation, as there 0? Stephan Original-Nachricht Datum: Sun, 10 Mar 2013 14:34:52 -0500 Von: Hyuntae Na h...@hotmail.com An: gmx-users@gromacs.org gmx-users@gromacs.org Betreff: [gmx-users] only 4 zero eigvals for NMA (with 10^-6 emtol for l-bfgs) Dear, This is the same issue that I asked last time. I wonder (1) if I setup the test wrong, (2) if NMA *can* have smaller than 6 eigenvalues even though the conformation of a protein is in equilibrium in the enough degree, or (2) if it gromacs does not make it in equilibrium because of the limit of the tolerance that gromacs can take care of. Is there anyone who can give me an advice for this? I updated the sample location in the case that the previous Skype link is not available. I am studying several protein fluctuation proteins using NMA (Normal Mode Analysis). In the mean time, I found out in many cases that NMA hessian matrix has about 3-4 zero eigenvalues even though having 6 zero eigenvalues are expected. I tried to reduce the tolerance level lesser than 10^-6 for the l-bfgs minimization, but still it have only 4 zero eigenvalues. Following is the 9 smallest eigenvalues (you can see that it has only 4 close-to-zero eigenvalues): -1.8476593e-06-1.2164109e-06-1.4830436e-071.3104790e-061.1364704e+001.7418209e+002.2836573e+003.7218206e+004.4875873e+00 Theoretically, NMA should have 6 zero eigenvalues which represents the freedom of the rotation and translation. I wonder why this happens. Is there anyone who can help me for this problem? In order to specify the case, I leave a link to download a zip file that contains files for one NMA tests having the problem: script, mdp, pdb, gro, topol, etc. The zip file also includes the eigenvalue/vector files. http://129.186.69.109/NMA-test.zip Thank you very much. -- Hyuntae -- gmx-users mailing listgmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
[gmx-users] removing rotational and translational motion (holonomic contraint)
Hi guys I am running a set of umbrella sampling simulations using gromacs 333 and the pull code. The idea is to apply an umbrella potential to restrain the distance between a protein and a ligand in the x direction. Unfortunately, the rotational and translational movement of the protein during the MD run interferes with my restraining of the protein-ligand distance. I am therefore interested in applying a rotational and translational constraint to my protein. Amadei et al described a solution to this problem in their J Chem Phys paper (see below). The problem was also discussed in the following gromacs post: http://lists.gromacs.org/pipermail/gmx-users/2003-May/005591.html Extract from the post: For a simulation in solvent it is less trivial but feasible as was shown by Andrea Amadei and coworkers. They devised a method which constrains roto-translational movements of a solute, described in J. Chem. Phys. 112(1) pp. 9-23. They've implemented the method in Gromacs and a patch can be obtained from Danilo Roccatano ( roccata at caspur dot it ) from that group. We will be trying to implement things properly into Gromacs in the near future here in Groningen, but it has to wait a bit. If it's done we'll put it forward for a future release. There is another post in 2006 that refers to a patch for removing roto-translational motion but the link no longer works. http://gromacs.5086.n6.nabble.com/Protein-rotation-in-triclinic-waterbox-td4413734.htmhttp://gromacs.5086.n6.nabble.com/Protein-rotation-in-triclinic-waterbox-td4413734.html Finally, there is a more recent post (2012) stating it has been implemented in version 4.5.1 and 4.5.3 but not released. http://gromacs.5086.n6.nabble.com/roto-translational-constraints-td4390750.html Has this feature been implemented in the official gromacs version? Can I obtain the patch and use it with gromacs 333? Thanks a lot Evelyne -- gmx-users mailing listgmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
Re: [gmx-users] query for gromacs-4.5.4
Hello Sir, Actually I have been given work to setup gromacs-4.5.4 in our cluster which is being used by users.I am not gromacs user and not aware of its internal details. I have got only .tpr file from user and I need to test my installation using that .tpr file. It works fine for 2 nodes 8 processes , 1 node 8 processes. But when number of processes are equal to 16 it gives segmentation fault and if number of processes are equal to 32 it gives error message like HYD_pmcd_pmiserv_send_signal (./pm/pmiserv/pmiserv_cb.c:221): assert (!closed) failed ui_cmd_cb (./pm/pmiserv/pmiserv_pmci.c:128): unable to send SIGUSR1 downstream HYDT_dmxu_poll_wait_for_event (./tools/demux/demux_poll.c:77): callback returned error status HYD_pmci_wait_for_completion (./pm/pmiserv/pmiserv_pmci.c:388): error waiting for event [ main (./ui/mpich/mpiexec.c:718): process manager error waiting for completion I am not clear like whether problem is there in my installation or what? Thanks and Regards, Chaitalij On Wed, Mar 6, 2013 at 5:41 PM, Justin Lemkul jalem...@vt.edu wrote: On 3/6/13 4:20 AM, Chaitali Chandratre wrote: Dear Sir , I am new to this installation and setup area. I need some information for -stepout option for What more information do you need? mdrun_mpi and also probable causes for segmentation fault in gromacs-4.5.4. (my node having 64 GB mem running with 16 processes, nsteps = 2000) There are too many causes to name. Please consult http://www.gromacs.org/ **Documentation/Terminology/**Blowing_Uphttp://www.gromacs.org/Documentation/Terminology/Blowing_Up. If you need further help, please be more specific, including a description of the system, steps taken to minimize and/or equilibrate it, and any complete .mdp file(s) that you are using. -Justin -- ==**== Justin A. Lemkul, Ph.D. Research Scientist Department of Biochemistry Virginia Tech Blacksburg, VA jalemkul[at]vt.edu | (540) 231-9080 http://www.bevanlab.biochem.**vt.edu/Pages/Personal/justinhttp://www.bevanlab.biochem.vt.edu/Pages/Personal/justin ==**== -- gmx-users mailing listgmx-users@gromacs.org http://lists.gromacs.org/**mailman/listinfo/gmx-usershttp://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/** Support/Mailing_Lists/Searchhttp://www.gromacs.org/Support/Mailing_Lists/Searchbefore posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. * Can't post? Read http://www.gromacs.org/**Support/Mailing_Listshttp://www.gromacs.org/Support/Mailing_Lists -- gmx-users mailing listgmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
