LAMMPS安装

加载CUDA环境

export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH

加载intel编译环境

source /opt/intel/impi/4.1.1.036/bin64/mpivars.sh
source /opt/intel/composer_xe_2013_sp1.0.080/bin/compilervars.sh intel64
source /opt/intel/composer_xe_2013_sp1.0.080/mkl/bin/mklvars.sh intel64	

编译LAMMPS的GPU库

解压lammps安装包，进入lib/gpu

make -f Makefile.linux	

复制编译文件

进入src/MAKE/OPTIONS，

cp Makefile.intel_cpu_intelmpi ../	

编译LAMMPS

进入src，

make yes-asphere yes-kspace yes-gpu
make yes-asphere yes-class2 yes-kspace yes-manybody yes-misc yes-molecule
make yes-mpiio yes-opt yes-replica yes-rigid
make yes-user-omp yes-user-intel
make intel_cpu_intelmpi(如果出现错误，再次执行该命令)

运行LAMMPS

mpirun -np NCPU where/is/lmp_intel_cpu_intelmpi -sf gpu -pk gpu NGPU -i file.in -o file.out

配置job.pbs

#PBS -N lammps      
#PBS -l nodes=1:ppn=NGPU
#PBS -q test               
#PBS -l walltime=24:00:00

export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
source /opt/intel/impi/4.1.1.036/bin64/mpivars.sh
source /opt/intel/composer_xe_2013_sp1.0.080/bin/compilervars.sh intel64
source /opt/intel/composer_xe_2013_sp1.0.080/mkl/bin/mklvars.sh intel64

cd $PBS_O_WORKDIR

mpirun -np NCPU where/is/lmp_intel_cpu_intelmpi -sf gpu -pk gpu NGPU -i file.in -o file.out