Skip to content

build-mlperf-inference-server-nvidia

Automatically generated README for this automation recipe: build-mlperf-inference-server-nvidia

Category: MLPerf benchmark support

License: Apache 2.0

  • Notes from the authors, contributors and users: README-extra

  • CM meta description for this script: _cm.yaml

  • Output cached? True

Reuse this script in your project

Install MLCommons CM automation meta-framework

Pull CM repository with this automation recipe (CM script)

cm pull repo mlcommons@cm4mlops

cmr "build mlcommons mlperf inference inference-server server nvidia-harness nvidia" --help

Run this script

Run this script via CLI
cm run script --tags=build,mlcommons,mlperf,inference,inference-server,server,nvidia-harness,nvidia[,variations] [--input_flags]
Run this script via CLI (alternative)
cmr "build mlcommons mlperf inference inference-server server nvidia-harness nvidia [variations]" [--input_flags]
Run this script from Python
import cmind

r = cmind.access({'action':'run'
              'automation':'script',
              'tags':'build,mlcommons,mlperf,inference,inference-server,server,nvidia-harness,nvidia'
              'out':'con',
              ...
              (other input keys for this script)
              ...
             })

if r['return']>0:
    print (r['error'])
Run this script via Docker (beta)
cm docker script "build mlcommons mlperf inference inference-server server nvidia-harness nvidia[variations]" [--input_flags]

Variations

  • Group "code"

    Click here to expand this section.

    • _ctuning (default)
    • _custom
    • _go
    • _mlcommons
    • _nvidia-only

  • Group "device"

    Click here to expand this section.

    • _cpu
      • ENV variables:
        • CM_MLPERF_DEVICE: cpu
    • _cuda (default)
      • ENV variables:
        • CM_MLPERF_DEVICE: cuda
        • CM_MLPERF_DEVICE_LIB_NAMESPEC: cudart
    • _inferentia
      • ENV variables:
        • CM_MLPERF_DEVICE: inferentia

  • Group "version"

    Click here to expand this section.

    • _r4.0
Default variations

_ctuning,_cuda

Script flags mapped to environment

  • --clean=value → CM_MAKE_CLEAN=value
  • --custom_system=value → CM_CUSTOM_SYSTEM_NVIDIA=value

Default environment

These keys can be updated via --env.KEY=VALUE or env dictionary in @input.json or using script flags.

  • CM_MAKE_BUILD_COMMAND: build
  • CM_MAKE_CLEAN: no
  • CM_CUSTOM_SYSTEM_NVIDIA: yes

Versions

Default version: r3.1

  • r2.1
  • r3.0
  • r3.1
  • r4.0

Native script being run

No run file exists for Windows

Script output

cmr "build mlcommons mlperf inference inference-server server nvidia-harness nvidia [variations]" [--input_flags] -j