|
最近实验室同时用上了深圳超算和天津超算。其中深圳超算的VASP是我自己编译的 天津超算的VASP是用的别人的。然后发现自己编译的VASP跟天津超算的相比速度差距特别大(肯定不是硬件造成的差距)。自己对VASP编译不怎么懂,只是能勉强编译通过,求教编译时候对makefile还有数据库什么的应该做些什么优化呢? 各种求指教!!!!
PS 深超用的为openmpi-1.4.4, intel/composer_xe_2011_sp1.7.256
PPS 我在我们自己的集群上用openmpi-1.8和 intel/composer_xe_2013.3.163,编译出来的计算速度和深超上也差不多。
附上深圳超算的Makefile 。
===========================================
.SUFFIXES: .inc .f .f90 .F
SUFFIX=.f90
#-----------------------------------------------------------------------
FC=ifort
# fortran linker
FCL=$(FC)
CPP_ = ./preprocess <$*.F | /usr/bin/cpp -P -C -traditional >$*$(SUFFIX)
#-----------------------------------------------------------------------
CPP_=fpp -f_com=no -free -w0 $*.F $*$(SUFFIX)
# -DRPROMU_DGEMV -DRACCMU_DGEMV
OFLAG=-O2 -ip -heap-arrays 64
OFLAG_HIGH = $(OFLAG)
OBJ_HIGH =
OBJ_NOOPT =
DEBUG = -FR -O0
INLINE = $(OFLAG)
LIB = -L../vasp.5.lib -ldmy \
../vasp.5.lib/linpack_double.o $(LAPACK) \
$(BLAS)
# options for linking (for compiler version 6.X) nothing is required
LINK =
#-----------------------------------------------------------------------
# since this version is faster on P4 machines, we recommend to use it
#-----------------------------------------------------------------------
FFT3D = fft3dfurth.o fft3dlib.o
#=======================================================================
# MPI section
#
# the system we used is an SGI test system, and it is best
# to compile using ifort and adding the option -lmpi during
# linking
#=======================================================================
FC=mpif90
FCL=$(FC)
#-----------------------------------------------------------------------
# additional options for CPP in parallel version (see also above):
# NGZhalf charge density reduced in Z direction
# wNGZhalf gamma point only reduced in Z direction
# scaLAPACK use scaLAPACK (usually slower on 100 Mbit Net)
#-----------------------------------------------------------------------
CPP = $(CPP_) -DMPI -DHOST=\"LinuxIFCmkl\" -DIFC \
-DCACHE_SIZE=32000 -DPGF90 -Davoidalloc -DNGZhalf \
-DMPI_BLOCK=8000 \
## -DRPROMU_DGEMV -DRACCMU_DGEMV
#-----------------------------------------------------------------------
#-----------------------------------------------------------------------
Makefile
(13.15 KB, 下载次数 Times of downloads: 11)
#BLACS=$(HOME)/archives/SCALAPACK/BLACS/
#SCA_=$(HOME)/archives/SCALAPACK/SCALAPACK
#SCA= $(SCA_)/libscalapack.a \
#SCA=/home-gg/compiler/intel/composer_xe_2011_sp1.7.256/mkl/lib/intel64/libmkl_scalapack_ilp64.a
SCA=
#-----------------------------------------------------------------------
# libraries for mpi
#-----------------------------------------------------------------------
LIB = -L../vasp.5.lib -ldmy \
../vasp.5.lib/linpack_double.o $(LAPACK) \
$(SCA) $(BLAS) \
# FFT: fftmpi.o with fft3dlib of Juergen Furthmueller
#FFT3D = fftmpi.o fftmpi_map.o fft3dfurth.o fft3dlib.o
# fftw.3.1.2 is slighly faster and should be used if available
#-----------------------------------------------------------------------
# general rules and compile lines
#-----------------------------------------------------------------------
BASIC= symmetry.o symlib.o lattlib.o random.o
|
|