Skip to content

Latest commit

 

History

History
191 lines (153 loc) · 7.25 KB

10_AMBER20.md

File metadata and controls

191 lines (153 loc) · 7.25 KB
Raw

Amber is a suite of biomolecular simulation programs. It began in the late 1970's, and is maintained by an active development community.

The term "Amber" refers to two things. First, it is a set of molecular mechanical force fields for the simulation of biomolecules (these force fields are in the public domain, and are used in a variety of simulation programs). Second, it is a package of molecular simulation programs which includes source code and demos.

Amber is distributed in two parts: AmberTools20 and Amber20. You can use AmberTools20 without Amber20, but not vice versa.

Ref: https://ambermd.org/

Pre-installation

System: CentOS Linux release 7.6.1810 with CMake 3.20.1, CUDA 11.1.1, GNU10 v10.3.0, OpenMPI4 v4.1.1 operated by Lmod

Refs:
https://www.hull1.com/linux/2020/08/21/complie-amber20.html
https://ambermd.org/doc12/Amber20.pdf

Install Dependencies

yum -y install tcsh make \
			   gcc gcc-gfortran gcc-c++ \
			   which flex bison patch bc \
			   libXt-devel libXext-devel \
			   perl perl-ExtUtils-MakeMaker util-linux wget \
			   bzip2 bzip2-devel zlib-devel tar

Extract amber

tar xvfj AmberTools20.tar.bz2
tar xvfj Amber20.tar.bz2

Create amber20 folder at destination path (e.g. /apps/amber20)

mkdir /apps/amber20

Upgrade and update Amber

cd amber20_src
./update_amber --upgrade
./update_amber --update

Compile Serial CPU

Complie and install

cd amber20_src
cd build
cmake .. -DCMAKE_INSTALL_PREFIX=/apps/amber20 -DCOMPILER=GNU -DCUDA=FALSE -DINSTALL_TESTS=FALSE -DDOWNLOAD_MINICONDA=TRUE -DMINICONDA_USE_PY3=TRUE
make install -j8

Test

cd /apps/amber20
source amber.sh
make test.serial

Compile for Serial GPU

Ref: https://ambermd.org/GPUHardware.php

Amber tries to support all CUDA SDK versions up to 11.x. In the past, they have recommended CUDA 9.1 or 9.2 for the best speed of the resulting executables, but this needs to be revisited. Here are the minimum requirements for different tiers of hardware:

  • Ampere (SM_80) based cards require CUDA 11.0 or later (RTX-3060, RTX-3070, RTX-3080, RTX-3090, RTX-A6000, A100).

  • Turing (SM_75) based cards require CUDA 9.2 or later (RTX-2080Ti, RTX-2080, RTX-2070, Quadro RTX6000/5000/4000).

  • Volta (SM_70) based cards require CUDA 9.0 or later (Titan-V, V100, Quadro GV100).

  • Pascal (SM_61) based cards require CUDA 8.0 or later. (Titan-XP [aka Pascal Titan-X], GTX-1080TI / 1080 / 1070 / 1060, Quadro P6000 / P5000, P4 / P40).

  • GTX-1080 cards require NVIDIA Driver version >= 367.27 for reliable numerical results.

  • GTX-Titan and GTX-780 cards require NVIDIA Driver version >= 319.60 for correct numerical results.

  • GTX-780Ti cards require a modified Bios from Exxact Corp to give correct numerical results.

  • GTX-Titan-Black Edition cards require NVIDIA Driver version >= 337.09 or 331.79 or later for correct numerical results.

Complie and install

cd amber20_src
cd build
make clean
cmake .. -DCMAKE_INSTALL_PREFIX=/apps/amber20 -DCOMPILER=GNU -DCUDA=TRUE -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-9.2 -DINSTALL_TESTS=FALSE -DDOWNLOAD_MINICONDA=TRUE -DMINICONDA_USE_PY3=TRUE
make install -j8

Test

cd /apps/amber20
source amber.sh
make test.cuda

Compile for Parallel CPU

Complie and install

cd amber20_src
cd build
make clean
cmake .. -DCMAKE_INSTALL_PREFIX=/apps/amber20 -DCOMPILER=GNU -DCUDA=FALSE -DMPI=TRUE -DINSTALL_TESTS=FALSE -DDOWNLOAD_MINICONDA=TRUE -DMINICONDA_USE_PY3=TRUE
make install -j8

Test

cd /apps/amber20
export DO_PARALLEL='mpirun -np 2'
test -f amber.sh  && source amber.sh
export DO_PARALLEL='mpirun -np 2'
make test.parallel

Compile for Parallel GPU

Ref: https://ambermd.org/GPUHardware.php

Amber tries to support all CUDA SDK versions up to 11.x. In the past, they have recommended CUDA 9.1 or 9.2 for the best speed of the resulting executables, but this needs to be revisited. Here are the minimum requirements for different tiers of hardware:

  • Ampere (SM_80) based cards require CUDA 11.0 or later (RTX-3060, RTX-3070, RTX-3080, RTX-3090, RTX-A6000, A100).

  • Turing (SM_75) based cards require CUDA 9.2 or later (RTX-2080Ti, RTX-2080, RTX-2070, Quadro RTX6000/5000/4000).

  • Volta (SM_70) based cards require CUDA 9.0 or later (Titan-V, V100, Quadro GV100).

  • Pascal (SM_61) based cards require CUDA 8.0 or later. (Titan-XP [aka Pascal Titan-X], GTX-1080TI / 1080 / 1070 / 1060, Quadro P6000 / P5000, P4 / P40).

  • GTX-1080 cards require NVIDIA Driver version >= 367.27 for reliable numerical results.

  • GTX-Titan and GTX-780 cards require NVIDIA Driver version >= 319.60 for correct numerical results.

  • GTX-780Ti cards require a modified Bios from Exxact Corp to give correct numerical results.

  • GTX-Titan-Black Edition cards require NVIDIA Driver version >= 337.09 or 331.79 or later for correct numerical results.

Complie and install

cd amber20_src
cd build
make clean
cmake .. -DCMAKE_INSTALL_PREFIX=/apps/amber20 -DCOMPILER=GNU -DMPI=TRUE -DCUDA=TRUE -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-9.2 -DINSTALL_TESTS=FALSE -DDOWNLOAD_MINICONDA=TRUE -DMINICONDA_USE_PY3=TRUE
make install -j8

Test

cd /apps/amber20
export DO_PARALLEL='mpirun -np 2'
test -f amber.sh  && source amber.sh
export DO_PARALLEL='mpirun -np 2'
make test.cuda.parallel

Compile for Parallel GPU with NCCL

Ref: https://ambermd.org/GPUHardware.php

Amber tries to support all CUDA SDK versions up to 11.x. In the past, they have recommended CUDA 9.1 or 9.2 for the best speed of the resulting executables, but this needs to be revisited. Here are the minimum requirements for different tiers of hardware:

  • Ampere (SM_80) based cards require CUDA 11.0 or later (RTX-3060, RTX-3070, RTX-3080, RTX-3090, RTX-A6000, A100).

  • Turing (SM_75) based cards require CUDA 9.2 or later (RTX-2080Ti, RTX-2080, RTX-2070, Quadro RTX6000/5000/4000).

  • Volta (SM_70) based cards require CUDA 9.0 or later (Titan-V, V100, Quadro GV100).

  • Pascal (SM_61) based cards require CUDA 8.0 or later. (Titan-XP [aka Pascal Titan-X], GTX-1080TI / 1080 / 1070 / 1060, Quadro P6000 / P5000, P4 / P40).

  • GTX-1080 cards require NVIDIA Driver version >= 367.27 for reliable numerical results.

  • GTX-Titan and GTX-780 cards require NVIDIA Driver version >= 319.60 for correct numerical results.

  • GTX-780Ti cards require a modified Bios from Exxact Corp to give correct numerical results.

  • GTX-Titan-Black Edition cards require NVIDIA Driver version >= 337.09 or 331.79 or later for correct numerical results.

Download NVIDIA NCCL from https://developer.nvidia.com/nccl

Install PnetCDF

autoreconf -i
./configure --prefix=/apps/PnetCDF
make -j8
make install

Complie and install

cd amber20_src
export NCCL_HOME="~/apps/lib/nccl_2.6.4-1+cuda10.0_x86_64"
cd build
make clean
cmake .. -DCMAKE_INSTALL_PREFIX=/apps/amber20 -DCOMPILER=GNU -DMPI=TRUE -DCUDA=TRUE -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-9.2 -DNCCL=TRUE -DPnetCDF_C_LIBRARY=~/apps/pnetcdf/lib/ -DPnetCDF_C_INCLUDE_DIR=~/apps/pnetcdf/include/ -DINSTALL_TESTS=FALSE -DDOWNLOAD_MINICONDA=TRUE -DMINICONDA_USE_PY3=TRUE
make install -j8

Test

cd /apps/amber20
export DO_PARALLEL='mpirun -np 2'
test -f amber.sh  && source amber.sh
export DO_PARALLEL='mpirun -np 2'
make test.cuda.parallel