User Tools

Site Tools


cluster:161

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
cluster:161 [2018/09/20 09:31]
hmeij07
cluster:161 [2020/07/16 13:16] (current)
hmeij07
Line 4: Line 4:
 ==== lammps-11Aug17 ==== ==== lammps-11Aug17 ====
  
-ok, new version much easier+lammps-11Aug17 (n78)\\ 
 +and now lammps-22Aug18 (n33-n37)\\ 
 +and now lammps-5Jun19 (microway)\\
  
 Update:  Update: 
-  * n78/gtx1080 lammps 11aug17 (centos7, mpich3/mpic++, cuda 8/sm_61) +  * n78/gtx1080 lammps 11aug17 (centos7, mpich3/mpic++, cuda 8/sm_61, /usr/local) 
-  * n37/k20 lammps 22aug18 (centos7, openmpi 1.8.4/mpic++, cuda 9.2/sm_35)  +      * [[cluster:164|GTX 1080 Ti]] 
-  * n33/k20 lammsp 22aug18 (centos6, mvapich2/mpicxx, cuda 5/sm35)+  * n37/k20 lammps 22aug18 (centos7, openmpi 1.8.4/mpic++, cuda 9.2/sm_35, /usr/local 
 +      * [[cluster:173|K20 Redo Usage]] 
 +  * n33/k20 lammps 11aug17 + 22aug18 (centos6, openmpi 1.8.4, cuda 5/sm35, /share/apps/CENTOS6) 
 +      * this page 
 +  * n33/k20 older lammps (centos6, mvapich, cuda 5/sm_??, /cm/shared/apps) 
 +      * /home/hmeij/sharptail/run.gpu or gromacs.sub files 
  --- //[[hmeij@wesleyan.edu|Henk]] 2018/09/20 09:30//  --- //[[hmeij@wesleyan.edu|Henk]] 2018/09/20 09:30//
  
 <code> <code>
 +N78
 # devel packages for libjpeg, blas, lapack, boost # devel packages for libjpeg, blas, lapack, boost
  
Line 26: Line 34:
 # edits # edits
 CUDA_HOME = /usr/local/cuda CUDA_HOME = /usr/local/cuda
-# best fitting ARCH for GTX 1080 with CUDA 8+ 
 +# best fitting ARCH for GTX 1080 with CUDA 8+
 CUDA_ARCH = -arch=sm_61 CUDA_ARCH = -arch=sm_61
 +# best fitting ARCH for P100 with CUDA 8+
 +CUDA_ARCH = -arch=sm_60
 +# best fitting ARCH for RTX/T4 with CUDA 10+
 +CUDA_ARCH = -arch=sm_75
 http://arnon.dk/matching-sm-architectures-arch-and-gencode-for-various-nvidia-cards/ http://arnon.dk/matching-sm-architectures-arch-and-gencode-for-various-nvidia-cards/
-for GTX1080 on n78+ 
 +# generate DOUBLE_DOUBLE (case matters)
 CUDA_PRECISION = -D_DOUBLE_DOUBLE CUDA_PRECISION = -D_DOUBLE_DOUBLE
 make -f Makefile.linux.double make -f Makefile.linux.double
Line 43: Line 57:
 JPG_LIB =       -ljpeg JPG_LIB =       -ljpeg
  
 +
 +make purge
 make yes-rigid make yes-rigid
 make yes-gpu make yes-gpu
Line 50: Line 66:
 make yes-misc make yes-misc
 make yes-molecule make yes-molecule
 +make yes-user-omp
 +make package-update
 +
 which mpicxx which mpicxx
 make mpi make mpi
Line 55: Line 74:
  
 # repeat from scratch for SINGLE_SINGLE and SINGLE_DOUBLE # repeat from scratch for SINGLE_SINGLE and SINGLE_DOUBLE
 +# different makefiles same edits
  
  
Line 74: Line 94:
 <code> <code>
  
-on n33, a gpu node+N33, a gpu node
  
-  mkdir /tmp/laamps; cd !$+  mkdir /tmp/lammps; cd !$
   gunzip lammps-stable.tar.gz    gunzip lammps-stable.tar.gz 
   tar xvf lammps-stable.tar    tar xvf lammps-stable.tar 
Line 108: Line 128:
    cd ..    cd ..
    #no cp MAKE/Makefile.serial MAKE/Makefile.serial_icc      #no cp MAKE/Makefile.serial MAKE/Makefile.serial_icc  
-   # edit MAKE/MAkefile.serial+    
 +   add compiler flag (centos7, 2020: -std=c++11) 
 +    
 +   # edit MAKE/Makefile.serial
    # leave gcc++  icc/icpc -DLAMMPS_JPEG for LMP_INC    # leave gcc++  icc/icpc -DLAMMPS_JPEG for LMP_INC
    # JPG_INC =   -I/usr/include    # JPG_INC =   -I/usr/include
Line 189: Line 212:
 # need both /cm/shared... and /cm/local ... # need both /cm/shared... and /cm/local ...
  
-# the malaysan approach also works, see install file+# the malaysian approach also works, see install file
  
  
cluster/161.1537450283.txt.gz · Last modified: 2018/09/20 09:31 by hmeij07