moebiusband/MD-Bench
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
49 Commits 8 Branches 0 Tags 3.4 MiB
C 88.8%
Cuda 8.4%
Makefile 2.8%
d0d2bf8a0c
Go to file
Rafael Ravedutti d0d2bf8a0c Add OSACA output for SOA 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-27 00:36:34 +02:00
arch_analysis Add OSACA output for SOA 2021-04-27 00:36:34 +02:00
asm Add comments for second kernel variant on Intel AOS assembly 2021-04-16 16:31:27 +02:00
scripts Add script to get stub results for several configurations 2021-04-22 20:24:50 +02:00
src Show cycles per atom and neighbor 2021-04-22 16:50:22 +02:00
util Switch License to LGPL3 2020-08-19 10:47:40 +02:00
.gitignore Initial commit 2020-07-23 16:41:14 +02:00
config.mk Merge master branch into stub 2021-04-15 20:12:36 +02:00
include_CLANG.mk Streamline build system 2021-03-24 08:43:44 +01:00
include_GCC.mk Restructure timing and instrumentation. Add performance metric. 2021-04-15 14:55:02 +02:00
include_ICC.mk Restructure timing and instrumentation. Add performance metric. 2021-04-15 14:55:02 +02:00
LICENSE Switch License to LGPL3 2020-08-19 10:47:40 +02:00
Makefile Restructure timing and instrumentation. Add performance metric. 2021-04-15 14:55:02 +02:00
README.md Streamline build system 2021-03-24 08:43:44 +01:00

README.md

MD-Bench

A simple, sequential C implementation of the Mantevo miniMD benchmark in less than 1000 LOC.

Build

  1. Open config.mk and edit the TAG value according to the tool chain used. Currently supported is GCC, CLANG (LLVM), and ICC (Intel).
  2. Change DATA_LAYOUT and DATA_TYPE if desired in config.mk.
  3. Open and adapt the compiler flags in <include_<TOOLCHAIN>.mk, e.g. in include_ICC.mk for the Intel tool chain.
  4. Build the binary calling make.

You can clean intermediate build results with make clean, and all build results with make distclean. You have to call make clean before make if you changed the build settings.

Configuration

Currently all settings apart from the options described below are hard-coded in main.c.

Run the benchmark

Without any options 200 steps with system size 32x32x32 is used.

The default can be changed using the following options:

-n / --nsteps <int>:  set number of timesteps for simulation
-nx/-ny/-nz <int>:    set linear dimension of systembox in x/y/z direction