URL: http://www.ks.uiuc.edu/Research/namd/

Source URL: git clone https://charm.cs.illinois.edu/gerrit/namd

Categories: application, open-source

MD; Parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems.

Versions

NAMD 2.12 Arm 18 - NAMD 2.12 Arm 18 MPI -

Build details for version 2.12 with Arm Compiler for HPC 18

Configuration

NAMD version 2.12
CHARM++ version 6.8.0
Arm compiler version 18
FFTW3
OpenMPI version 3.0.0
Last updated 17/1/18

Build instructions

Build the dependencies

This recipe for building NAMD requires CHARM++ version 6.8.0 built with CHARMARCH=multicore-linux-aarch64.

Make a note of the install location as this will be required for the NAMD build. Choose install and build locations for NAMD, and set as environment variables INSTALL_DIR and BUILD_DIR respectively.

Download and unpack

cd $BUILD_DIR
wget http://www.ks.uiuc.edu/Research/namd/2.12/download/832164/NAMD_2.12_Source.tar.gz
tar -xf NAMD_2.12_Source.tar.gz

Compiler configuration

: ${CC:=armclang}
: ${CXX:=armclang++}
: ${FC:=armflang}
export CC CXX FC

Set tuning flags and (optionally) disable some compiler warnings:

TUNE="-march=armv8.1-a -mcpu=native"
WARN="-Wno-parentheses-equality -Wno-writable-strings -Wno-return-type -Wno-logical-op-parentheses -Wno-writable-strings -Wno-deprecated-register "

Build configuration

cd ./NAMD_2.12_Source

Generate an armclang architecture file based on the existing Linux-ARM64-g++.arch file. This will need to know the location of the CHARM build NAMD_CHARM_DIR, and the architecture used for the CHARM build CHARMARCH.

export CHARMARCH=multicore-linux-aarch64
export NAMD_CHARM_DIR=${INSTALL_DIR}/charm680

cp arch/Linux-ARM64-g++.arch arch/Linux-ARM64-armclang.arch
sed -i "s|CHARMBASE.*|CHARMBASE = ${NAMD_CHARM_DIR}/|g" Make.charm
sed -i "s|^CHARMARCH =.*|CHARMARCH = $CHARMARCH|g" arch/Linux-ARM64-armclang.arch
sed -i "s|^FLOATOPTS =.*|FLOATOPTS = -O3 -fomit-frame-pointer ${WARN} ${TUNE}|g" arch/Linux-ARM64-armclang.arch
sed -i "s|^CC = gcc|CC = $CC|g" arch/Linux-ARM64-armclang.arch
sed -i "s|^CXX = g++|CXX = $CXX |g" arch/Linux-ARM64-armclang.arch

Set the FFTW config based on the x86_64 one. This will need to link to the FFTW3 library, given by NAMD_FFTW3_DIR.

cp arch/Linux-x86_64.fftw3 arch/Linux-ARM64.fftw3
sed -i "s|^FFTDIR=.*|FFTDIR=$NAMD_FFTW3_DIR|g" arch/Linux-ARM64.fftw3
sed -i "s|^FFTINCL=.*|FFTINCL=-I$NAMD_FFTW3_DIR/include|g" arch/Linux-ARM64.fftw3
sed -i "s|^FFTLIB=.*|FFTLIB=-L$NAMD_FFTW3_DIR/lib -lfftw3|g" arch/Linux-ARM64.fftw3

Finally, configure, build...

./config Linux-ARM64-armclang --with-fftw3 --without-tcl --charm-arch $CHARMARCH
cd Linux-ARM64-armclang
make -j

...and create symbolic link to the executables.

cd $INSTALL_DIR
ln -s $BUILD_DIR/NAMD_2.12_Source/Linux-ARM64-armclang/charmrun charmrun
ln -s $BUILD_DIR/NAMD_2.12_Source/Linux-ARM64-armclang/namd2 namd2

Example decks

The Apoa1 example is a standard NAMD cross-platform benchmark and suitable for small platforms.

wget http://www.ks.uiuc.edu/Research/namd/utilities/apoa1.tar.gz

tar -xf apoa1.tar.gz

${INSTALL_DIR}/namd2 +p40 apoa1/apoa1.namd +setcpuaffinity

Build details for version 2.12 with Arm Compiler for HPC 18 MPI