Skip to content

/AWS1/IF_SGM=>CREATEOPTIMIZATIONJOB()

About CreateOptimizationJob

Creates a job that optimizes a model for inference performance. To create the job, you provide the location of a source model, and you provide the settings for the optimization techniques that you want the job to apply. When the job completes successfully, SageMaker uploads the new optimized model to the output destination that you specify.

For more information about how to use this action, and about the supported optimization techniques, see Optimize model inference with Amazon SageMaker.

Method Signature

METHODS /AWS1/IF_SGM~CREATEOPTIMIZATIONJOB
  IMPORTING
    !IV_OPTIMIZATIONJOBNAME TYPE /AWS1/SGMENTITYNAME OPTIONAL
    !IV_ROLEARN TYPE /AWS1/SGMROLEARN OPTIONAL
    !IO_MODELSOURCE TYPE REF TO /AWS1/CL_SGMOPTIMIZATIONJOBM00 OPTIONAL
    !IV_DEPLOYMENTINSTANCETYPE TYPE /AWS1/SGMOPTIMIZATIONJOBDEPL00 OPTIONAL
    !IV_MAXINSTANCECOUNT TYPE /AWS1/SGMOPTMZTNJOBMAXINSTCNT OPTIONAL
    !IT_OPTIMIZATIONENVIRONMENT TYPE /AWS1/CL_SGMOPTIMIZATIONJOBE00=>TT_OPTIMIZATIONJOBENVIRONMEN00 OPTIONAL
    !IT_OPTIMIZATIONCONFIGS TYPE /AWS1/CL_SGMOPTIMIZATIONCONFIG=>TT_OPTIMIZATIONCONFIGS OPTIONAL
    !IO_OUTPUTCONFIG TYPE REF TO /AWS1/CL_SGMOPTIMIZATIONJOBO00 OPTIONAL
    !IO_STOPPINGCONDITION TYPE REF TO /AWS1/CL_SGMSTOPPINGCONDITION OPTIONAL
    !IT_TAGS TYPE /AWS1/CL_SGMTAG=>TT_TAGLIST OPTIONAL
    !IO_VPCCONFIG TYPE REF TO /AWS1/CL_SGMOPTIMIZATIONVPCCFG OPTIONAL
  RETURNING
    VALUE(OO_OUTPUT) TYPE REF TO /aws1/cl_sgmcreoptimizationj01
  RAISING
    /AWS1/CX_SGMRESOURCEINUSE
    /AWS1/CX_SGMRESOURCELIMITEXCD
    /AWS1/CX_SGMCLIENTEXC
    /AWS1/CX_SGMSERVEREXC
    /AWS1/CX_RT_TECHNICAL_GENERIC
    /AWS1/CX_RT_SERVICE_GENERIC.

IMPORTING

Required arguments:

iv_optimizationjobname TYPE /AWS1/SGMENTITYNAME /AWS1/SGMENTITYNAME

A custom name for the new optimization job.

iv_rolearn TYPE /AWS1/SGMROLEARN /AWS1/SGMROLEARN

The Amazon Resource Name (ARN) of an IAM role that enables Amazon SageMaker AI to perform tasks on your behalf.

During model optimization, Amazon SageMaker AI needs your permission to:

  • Read input data from an S3 bucket

  • Write model artifacts to an S3 bucket

  • Write logs to Amazon CloudWatch Logs

  • Publish metrics to Amazon CloudWatch

You grant permissions for all of these tasks to an IAM role. To pass this role to Amazon SageMaker AI, the caller of this API must have the iam:PassRole permission. For more information, see Amazon SageMaker AI Roles.

io_modelsource TYPE REF TO /AWS1/CL_SGMOPTIMIZATIONJOBM00 /AWS1/CL_SGMOPTIMIZATIONJOBM00

The location of the source model to optimize with an optimization job.

iv_deploymentinstancetype TYPE /AWS1/SGMOPTIMIZATIONJOBDEPL00 /AWS1/SGMOPTIMIZATIONJOBDEPL00

The type of instance that hosts the optimized model that you create with the optimization job.

it_optimizationconfigs TYPE /AWS1/CL_SGMOPTIMIZATIONCONFIG=>TT_OPTIMIZATIONCONFIGS TT_OPTIMIZATIONCONFIGS

Settings for each of the optimization techniques that the job applies.

io_outputconfig TYPE REF TO /AWS1/CL_SGMOPTIMIZATIONJOBO00 /AWS1/CL_SGMOPTIMIZATIONJOBO00

Details for where to store the optimized model that you create with the optimization job.

io_stoppingcondition TYPE REF TO /AWS1/CL_SGMSTOPPINGCONDITION /AWS1/CL_SGMSTOPPINGCONDITION

Specifies a limit to how long a job can run. When the job reaches the time limit, SageMaker ends the job. Use this API to cap costs.

To stop a training job, SageMaker sends the algorithm the SIGTERM signal, which delays job termination for 120 seconds. Algorithms can use this 120-second window to save the model artifacts, so the results of training are not lost.

The training algorithms provided by SageMaker automatically save the intermediate results of a model training job when possible. This attempt to save artifacts is only a best effort case as model might not be in a state from which it can be saved. For example, if training has just started, the model might not be ready to save. When saved, this intermediate data is a valid model artifact. You can use it to create a model with CreateModel.

The Neural Topic Model (NTM) currently does not support saving intermediate model artifacts. When training NTMs, make sure that the maximum runtime is sufficient for the training job to complete.

Optional arguments:

iv_maxinstancecount TYPE /AWS1/SGMOPTMZTNJOBMAXINSTCNT /AWS1/SGMOPTMZTNJOBMAXINSTCNT

The maximum number of instances to use for the optimization job.

it_optimizationenvironment TYPE /AWS1/CL_SGMOPTIMIZATIONJOBE00=>TT_OPTIMIZATIONJOBENVIRONMEN00 TT_OPTIMIZATIONJOBENVIRONMEN00

The environment variables to set in the model container.

it_tags TYPE /AWS1/CL_SGMTAG=>TT_TAGLIST TT_TAGLIST

A list of key-value pairs associated with the optimization job. For more information, see Tagging Amazon Web Services resources in the Amazon Web Services General Reference Guide.

io_vpcconfig TYPE REF TO /AWS1/CL_SGMOPTIMIZATIONVPCCFG /AWS1/CL_SGMOPTIMIZATIONVPCCFG

A VPC in Amazon VPC that your optimized model has access to.

RETURNING

oo_output TYPE REF TO /aws1/cl_sgmcreoptimizationj01 /AWS1/CL_SGMCREOPTIMIZATIONJ01

Examples

Syntax Example

This is an example of the syntax for calling the method. It includes every possible argument and initializes every possible value. The data provided is not necessarily semantically accurate (for example the value "string" may be provided for something that is intended to be an instance ID, or in some cases two arguments may be mutually exclusive). The syntax shows the ABAP syntax for creating the various data structures.

DATA(lo_result) = lo_client->createoptimizationjob(
  io_modelsource = new /aws1/cl_sgmoptimizationjobm00(
    io_s3 = new /aws1/cl_sgmoptimizationjobm01(
      io_modelaccessconfig = new /aws1/cl_sgmoptimizationmdel00( ABAP_TRUE )
      iv_s3uri = |string|
    )
    io_sagemakermodel = new /aws1/cl_sgmoptmztnsmmodel( |string| )
  )
  io_outputconfig = new /aws1/cl_sgmoptimizationjobo00(
    io_sagemakermodel = new /aws1/cl_sgmoptmztnsmmodel( |string| )
    iv_kmskeyid = |string|
    iv_s3outputlocation = |string|
  )
  io_stoppingcondition = new /aws1/cl_sgmstoppingcondition(
    iv_maxpendingtimeinseconds = 123
    iv_maxruntimeinseconds = 123
    iv_maxwaittimeinseconds = 123
  )
  io_vpcconfig = new /aws1/cl_sgmoptimizationvpccfg(
    it_securitygroupids = VALUE /aws1/cl_sgmoptimizationvpcs00=>tt_optimizationvpcsecgroupids(
      ( new /aws1/cl_sgmoptimizationvpcs00( |string| ) )
    )
    it_subnets = VALUE /aws1/cl_sgmoptimizationvpcs01=>tt_optimizationvpcsubnets(
      ( new /aws1/cl_sgmoptimizationvpcs01( |string| ) )
    )
  )
  it_optimizationconfigs = VALUE /aws1/cl_sgmoptimizationconfig=>tt_optimizationconfigs(
    (
      new /aws1/cl_sgmoptimizationconfig(
        io_mdelspeculativedecoding00 = new /aws1/cl_sgmmdelspeculatived00(
          io_trainingdatasource = new /aws1/cl_sgmmdelspeculatived01(
            iv_s3datatype = |string|
            iv_s3uri = |string|
          )
          iv_technique = |string|
        )
        io_modelcompilationconfig = new /aws1/cl_sgmmdelcompilationcfg(
          it_overrideenvironment = VALUE /aws1/cl_sgmoptimizationjobe00=>tt_optimizationjobenvironmen00(
            (
              VALUE /aws1/cl_sgmoptimizationjobe00=>ts_optimizationjobenv00_maprow(
                key = |string|
                value = new /aws1/cl_sgmoptimizationjobe00( |string| )
              )
            )
          )
          iv_image = |string|
        )
        io_modelquantizationconfig = new /aws1/cl_sgmmdelquantization00(
          it_overrideenvironment = VALUE /aws1/cl_sgmoptimizationjobe00=>tt_optimizationjobenvironmen00(
            (
              VALUE /aws1/cl_sgmoptimizationjobe00=>ts_optimizationjobenv00_maprow(
                key = |string|
                value = new /aws1/cl_sgmoptimizationjobe00( |string| )
              )
            )
          )
          iv_image = |string|
        )
        io_modelshardingconfig = new /aws1/cl_sgmmodelshardingcfg(
          it_overrideenvironment = VALUE /aws1/cl_sgmoptimizationjobe00=>tt_optimizationjobenvironmen00(
            (
              VALUE /aws1/cl_sgmoptimizationjobe00=>ts_optimizationjobenv00_maprow(
                key = |string|
                value = new /aws1/cl_sgmoptimizationjobe00( |string| )
              )
            )
          )
          iv_image = |string|
        )
      )
    )
  )
  it_optimizationenvironment = VALUE /aws1/cl_sgmoptimizationjobe00=>tt_optimizationjobenvironmen00(
    (
      VALUE /aws1/cl_sgmoptimizationjobe00=>ts_optimizationjobenv00_maprow(
        key = |string|
        value = new /aws1/cl_sgmoptimizationjobe00( |string| )
      )
    )
  )
  it_tags = VALUE /aws1/cl_sgmtag=>tt_taglist(
    (
      new /aws1/cl_sgmtag(
        iv_key = |string|
        iv_value = |string|
      )
    )
  )
  iv_deploymentinstancetype = |string|
  iv_maxinstancecount = 123
  iv_optimizationjobname = |string|
  iv_rolearn = |string|
).

This is an example of reading all possible response values

lo_result = lo_result.
IF lo_result IS NOT INITIAL.
  lv_optimizationjobarn = lo_result->get_optimizationjobarn( ).
ENDIF.