Difference between revisions of "Benchmarking: HPL AMD optimised (High Performance Linpack)"

From Define Wiki
Jump to navigation Jump to search
Line 15: Line 15:
 
wget https://download.open-mpi.org/release/open-mpi/v2.1/openmpi-2.1.5.tar.gz
 
wget https://download.open-mpi.org/release/open-mpi/v2.1/openmpi-2.1.5.tar.gz
 
</syntaxhighlight>
 
</syntaxhighlight>
 +
 +
== Using AMD HPL ==
 +
 +
untar AMD_HPL
 +
 +
<syntaxhighlight>
 +
tar xvf AMD_HPL.tgz
 +
 +
</syntaxhighlight>
 +
 +
Change directory to "HPL" that will have been created from untar`ing.
 +
 +
Within HPL you will find a number of files but the most important ones will be:
 +
 +
"run_hpl_ccx.sh" and "HPL.dat"
 +
 +
within run_hpl_ccx.sh you may have to edit the absolute location of mpirun since it may not match your system.
 +
 +
open HPL.dat and edit the N value to be inline with your memory capacity.
 +
 +
Using HPL calulator to work out what this value should be relative to the system your testing.
  
 
== Getting HPL to loop for 3 runs (or more) ==
 
== Getting HPL to loop for 3 runs (or more) ==

Revision as of 14:23, 24 January 2019

HPL AMD Optimised

Getting it to work you have to compile openmpi 2.1.5 since it requires libs specific to that mpi version.

Where the tar file is stored

\\10.0.0.222\software\AMD\HPL\AMD_HPL.tgz

Dependencies required

To compile openmpi tools you will always need gcc etc, safe time with development tools.

yum group install "Development Tools"
wget https://download.open-mpi.org/release/open-mpi/v2.1/openmpi-2.1.5.tar.gz

Using AMD HPL

untar AMD_HPL

tar xvf AMD_HPL.tgz

Change directory to "HPL" that will have been created from untar`ing.

Within HPL you will find a number of files but the most important ones will be:

"run_hpl_ccx.sh" and "HPL.dat"

within run_hpl_ccx.sh you may have to edit the absolute location of mpirun since it may not match your system.

open HPL.dat and edit the N value to be inline with your memory capacity.

Using HPL calulator to work out what this value should be relative to the system your testing.

Getting HPL to loop for 3 runs (or more)

The absolute path to "run_hpl_ccx.sh" may change depending on where you untar AMD_HPL.tgz so just check that the path is correct here for you.

#! /bin/bash
for i in {1..3}; do /root/HPL/run_hpl_ccx.sh; done