jianmu
/
lammps
forked from lijiext/lammps
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #1066 from rbberger/doc-fixes 

minor tweak to docs
This commit is contained in:
Steve Plimpton 2018-08-21 08:01:30 -06:00 committed by GitHub
parent 2c19079707 2b5618dc40
commit ee3b7a67a0
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 34 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

54
doc/src/Speed_gpu.txt
View File

@ -9,17 +9,17 @@ Documentation"_ld - "LAMMPS Commands"_lc :c
GPU package :h3
The GPU package was developed by Mike Brown at ORNL and his
collaborators, particularly Trung Nguyen (ORNL). It provides GPU
versions of many pair styles, including the 3-body Stillinger-Weber
pair style, and for "kspace_style pppm"_kspace_style.html for
long-range Coulombics. It has the following general features:
The GPU package was developed by Mike Brown while at SNL and ORNL
and his collaborators, particularly Trung Nguyen (now at Northwestern).
It provides GPU versions of many pair styles and for parts of the
"kspace_style pppm"_kspace_style.html for long-range Coulombics.
It has the following general features:
It is designed to exploit common GPU hardware configurations where one
or more GPUs are coupled to many cores of one or more multi-core CPUs,
e.g. within a node of a parallel machine. :ulb,l
Atom-based data (e.g. coordinates, forces) moves back-and-forth
Atom-based data (e.g. coordinates, forces) are moved back-and-forth
between the CPU(s) and GPU every timestep. :l
Neighbor lists can be built on the CPU or on the GPU :l
@ -28,8 +28,8 @@ The charge assignment and force interpolation portions of PPPM can be
run on the GPU. The FFT portion, which requires MPI communication
between processors, runs on the CPU. :l
Asynchronous force computations can be performed simultaneously on the
CPU(s) and GPU. :l
Force computations of different style (pair vs. bond/angle/dihedral/improper)
can be performed concurrently on the GPU and CPU(s), respectively. :l
It allows for GPU computations to be performed in single or double
precision, or in mixed-mode precision, where pairwise forces are
@ -39,21 +39,32 @@ force vectors. :l
LAMMPS-specific code is in the GPU package. It makes calls to a
generic GPU library in the lib/gpu directory. This library provides
NVIDIA support as well as more general OpenCL support, so that the
same functionality can eventually be supported on a variety of GPU
hardware. :l
same functionality is supported on a variety of hardware. :l
:ule
[Required hardware/software:]
To use this package, you currently need to have an NVIDIA GPU and
install the NVIDIA CUDA software on your system:
To compile and use this package in CUDA mode, you currently need
to have an NVIDIA GPU and install the corresponding NVIDIA CUDA
toolkit software on your system (this is primarily tested on Linux
and completely unsupported on Windows):
Check if you have an NVIDIA GPU: cat
/proc/driver/nvidia/gpus/0/information Go to
http://www.nvidia.com/object/cuda_get.html Install a driver and
toolkit appropriate for your system (SDK is not necessary) Run
lammps/lib/gpu/nvc_get_devices (after building the GPU library, see
below) to list supported devices and properties :ul
Check if you have an NVIDIA GPU: cat /proc/driver/nvidia/gpus/*/information :ulb,l
Go to http://www.nvidia.com/object/cuda_get.html :l
Install a driver and toolkit appropriate for your system (SDK is not necessary) :l
Run lammps/lib/gpu/nvc_get_devices (after building the GPU library, see below) to
list supported devices and properties :ule,l
To compile and use this package in OpenCL mode, you currently need
to have the OpenCL headers and the (vendor neutral) OpenCL library installed.
In OpenCL mode, the acceleration depends on having an "OpenCL Installable Client
Driver (ICD)"_https://www.khronos.org/news/permalink/opencl-installable-client-driver-icd-loader
installed. There can be multiple of them for the same or different hardware
(GPUs, CPUs, Accelerators) installed at the same time. OpenCL refers to those
as 'platforms'. The GPU library will select the [first] suitable platform,
but this can be overridded using the device option of the "package"_package.html
command. run lammps/lib/gpu/ocl_get_devices to get a list of available
platforms and devices with a suitable ICD available.
[Building LAMMPS with the GPU package:]
@ -120,7 +131,10 @@ GPUs/node to use, as well as other options.
The performance of a GPU versus a multi-core CPU is a function of your
hardware, which pair style is used, the number of atoms/GPU, and the
precision used on the GPU (double, single, mixed).
precision used on the GPU (double, single, mixed). Using the GPU package
in OpenCL mode on CPUs (which uses vectorization and multithreading) is
usually resulting in inferior performance compared to using LAMMPS' native
threading and vectorization support in the USER-OMP and USER-INTEL packages.
See the "Benchmark page"_http://lammps.sandia.gov/bench.html of the
LAMMPS web site for performance of the GPU package on various
@ -146,7 +160,7 @@ The "package gpu"_package.html command has several options for tuning
performance. Neighbor lists can be built on the GPU or CPU. Force
calculations can be dynamically balanced across the CPU cores and
GPUs. GPU-specific settings can be made which can be optimized
for different hardware. See the "packakge"_package.html command
for different hardware. See the "package"_package.html command
doc page for details. :l
As described by the "package gpu"_package.html command, GPU