Running benchmark scripts

Documentation on how to run the benchmark scripts

Here's the way I run the benchmark scripts. I create a new directory where I'm going to put the job scripts and output files:

mkdir FPD_testing_results

cd FPD_testing_results

then I link the generator script out of this repo into that directory:

ln -s ../path/to/repo/parfu_archive_tool/benchmark_scripts/gen_FPD_test_script.bash

To configure the generator script, set these environment variables:

FPD_RUNTIME=24:00:00 (optional)

FPD_SINGLE_RANK_NODE=yes ("yes" for pigz, tar, tar_gz, "no" for everything else)

FPD_SYSTEM=bw_moab

FPD_DATASET=GW

FPD_CODE=tar_gz

[email protected]

then run it:

csteffen@h2ologin2 09:28 ~/FP_testing_tar_targz_2020feb $ ./gen_FPD_test_script.bash

script file name= >>> FPD_test_bw_moab_000006.bash <<<

RANKS=1

csteffen@h2ologin2 09:30 ~/FP_testing_tar_targz_2020feb $