Re: [gmx-users] Regarding Gromacs 5.0.3 parallel computation
Le 10/12/2014 07:22, Bikash Ranjan Sahoo a écrit : I tried to do a small simulation in Gromacs 4.5.5 using 30 cores for 200 ps. The computation time was 4.56 minutes . The command used was dplace -c 0-29 mdrun -v -s md.tpr -c md.gro -nt 30 . Next I ran the same system using Gromacs 5.0.3. The command used was dplace -c 0-29 mpirun -np 30 mdrun_mpi -v -s md.tpr -c md.gro. The simulation was extremely slow and took 37 minutes to complete only 200 ps MD. Dear Bikash, You are more or less benchmarking threads versus mpi, right? Note also that dplace for such rounded number (30) is probably not optimal either except if your number of processors is in the multiplicity of 30 (hexacores?). Best, Stéphane -- Team Protein Design In Silico UFIP, UMR 6286 CNRS, UFR Sciences et Techniques, 2, rue de la Houssinière, Bât. 25, 44322 Nantes cedex 03, France Tél : +33 251 125 636 Fax : +33 251 125 632 http://www.ufip.univ-nantes.fr/ - http://www.steletch.org -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Regarding Gromacs 5.0.3 parallel computation
Dear Dr. Stéphane, Thank you for your quick reply. How can I solve this? Can you please guide me in getting the right command to run mdrun as fast as gromacs 4.5.5 -nt command. I am pasting the architecture of my cluster below. Kindly help me understanding how can I modify the mdrun_mpi command to access multi cores in my cluster. Interactive Use 276 CPU Avail - Free Interactive Use CPU List: 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 193 194 195 196 197 198 199 203 209 217 218 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 System Topology (CPU position) -- column1 column2 column3 column4 row6 120-131 48-59 192-203 264-275 row5 108-119 36-47 180-191 252-263 row4 96-107 24-35 168-179 240-251 row3 84-9512-23 156-167 228-239 row2 72-83 0-11 144-155 216-227 row1 60-71 - 132-143 204-215 Memory Information -- Interactive Use Free/Total Memory: 1469 GB / 1514 GB Best regards In anticipation of your reply Bikash On Wed, Dec 10, 2014 at 5:38 PM, Téletchéa Stéphane stephane.teletc...@univ-nantes.fr wrote: Le 10/12/2014 07:22, Bikash Ranjan Sahoo a écrit : I tried to do a small simulation in Gromacs 4.5.5 using 30 cores for 200 ps. The computation time was 4.56 minutes . The command used was dplace -c 0-29 mdrun -v -s md.tpr -c md.gro -nt 30 . Next I ran the same system using Gromacs 5.0.3. The command used was dplace -c 0-29 mpirun -np 30 mdrun_mpi -v -s md.tpr -c md.gro. The simulation was extremely slow and took 37 minutes to complete only 200 ps MD. Dear Bikash, You are more or less benchmarking threads versus mpi, right? Note also that dplace for such rounded number (30) is probably not optimal either except if your number of processors is in the multiplicity of 30 (hexacores?). Best, Stéphane -- Team Protein Design In Silico UFIP, UMR 6286 CNRS, UFR Sciences et Techniques, 2, rue de la Houssinière, Bât. 25, 44322 Nantes cedex 03, France Tél : +33 251 125 636 Fax : +33 251 125 632 http://www.ufip.univ-nantes.fr/ - http://www.steletch.org -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Regarding Gromacs 5.0.3 parallel computation
Le 10/12/2014 12:28, Bikash Ranjan Sahoo a écrit : Dear Dr. Stéphane, Thank you for your quick reply. How can I solve this? Can you please guide me in getting the right command to run mdrun as fast as gromacs 4.5.5 -nt command. I am pasting the architecture of my cluster below. Kindly help me understanding how can I modify the mdrun_mpi command to access multi cores in my cluster. Dear Bikash, Why not using the -nt option in gromacs 5.0.3 too? My point was that you should use the same parameters for comparison of performance... What I would do is: mdrun -nt 30 -pin auto Try first using the gromacs tools (-pin auto) than using the dplace command since in your case you ask it to split you job on different cpu according to the diagram you showed. At least you could use the logical architecture for the dplace command: http://techpubs.sgi.com/library/tpl/cgi-bin/getdoc.cgi?coll=linuxdb=bkssrch=fname=/SGI_Developer/LX_AppTune/sgi_html/ch05.html Last, you should also try to first tune the pme decomposition using the tunepme option on a short run to see if the rather rough domain decomposition done by mdrun in a first approach is optimal for your system. You should also try to use cpu multiplicity and power of two for maximal performance, in your case, probably something like -nt 12, -nt 24 or -nt 36, since each cpu seems to be a 12-core ... See their respective manuals and command line helps for more info. Best, Stéphane -- Team Protein Design In Silico UFIP, UMR 6286 CNRS, UFR Sciences et Techniques, 2, rue de la Houssinière, Bât. 25, 44322 Nantes cedex 03, France Tél : +33 251 125 636 Fax : +33 251 125 632 http://www.ufip.univ-nantes.fr/ - http://www.steletch.org -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] Regarding Gromacs 5.0.3 parallel computation
Dear All, I have installed the Gromacs 5.0.3 in cluster and would like to thank Dr. Mark for his valuable suggestions and guidance. I am facing some problems in the computation speed in 5.0.3. A comparable study of the same system in gromacs 4.5.5 and 5.0.3 in the same cluster using equal number of nodes rendered an extremely slow simulation for the latter one. The commands I used for installation are pasted below. cmake .. -DCMAKE_INSTALL_PREFIX=/user1/GROMACS-5.0.3 -DGMX_MPI=ON -DGMX_THREAD_MPI=ON -DGMX_PREFER_STATIC_LIBS=ON -DGMX_BUILD_OWN_FFTW=ON -DGMX_X11=OFF -DGMX_CPU_ACCELERATION=SSE4.1 I tried to do a small simulation in Gromacs 4.5.5 using 30 cores for 200 ps. The computation time was 4.56 minutes . The command used was dplace -c 0-29 mdrun -v -s md.tpr -c md.gro -nt 30 . Next I ran the same system using Gromacs 5.0.3. The command used was dplace -c 0-29 mpirun -np 30 mdrun_mpi -v -s md.tpr -c md.gro. The simulation was extremely slow and took 37 minutes to complete only 200 ps MD. Even the energy minimization for a small protein is taking long time in 5.0.3 which can be converged in 4.5.5 in few seconds. Kindly suggest me where is the problem. Is there any problem in my installation procedure (in cmake commands). Thanking You In anticipation of your reply. Bikash, Osaka, Japan P.S. The dplace 0-29 is for serial assignment of CPUs in my cluster. Kindly ignore it if you are using qsub. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.