MLPerf: Difference between revisions

From HPCWIKI
Jump to navigation Jump to search
(Created page with "[https://mlcommons.org/en/ MLPerf] is a consortium of key contributors from the AI/ML (Artificial Intelligence and Machine Learning) community that provides unbiased AI/ML performance evaluations of hardware, software, and services. == References ==")
 
No edit summary
Line 1: Line 1:
[https://mlcommons.org/en/ MLPerf] is a consortium of key contributors from the AI/ML (Artificial Intelligence and Machine Learning) community that provides unbiased AI/ML performance evaluations of hardware, software, and services.
[https://mlcommons.org/en/ MLPerf] is a consortium of key contributors from the AI/ML (Artificial Intelligence and Machine Learning) community that provides unbiased AI/ML performance evaluations of hardware, software, and services.


== Benchmark Script ==
<syntaxhighlight lang="bash">
#MIG slice benchmark
trap "date; echo failed :(; exit 1" ERR # catch execution failures
ALL_GPUS=$(nvidia-smi -L | grep "UUID: MIG-GPU" | cut -d" " -f8 | cut -d')' -f1)
for gpu in $(echo "$ALL_GPUS"); do
    export CUDA_VISIBLE_DEVICES=$gpu
    $MLPERF_BENCHMARK & # launch workload in background
done
wait # wait for the completion of all the background processes
</syntaxhighlight>


== References ==
== References ==

Revision as of 11:09, 22 May 2023

MLPerf is a consortium of key contributors from the AI/ML (Artificial Intelligence and Machine Learning) community that provides unbiased AI/ML performance evaluations of hardware, software, and services.

Benchmark Script

#MIG slice benchmark
trap "date; echo failed :(; exit 1" ERR # catch execution failures
ALL_GPUS=$(nvidia-smi -L | grep "UUID: MIG-GPU" | cut -d" " -f8 | cut -d')' -f1)
for gpu in $(echo "$ALL_GPUS"); do
    export CUDA_VISIBLE_DEVICES=$gpu
    $MLPERF_BENCHMARK & # launch workload in background
done

wait # wait for the completion of all the background processes

References