aws-sdk-sagemaker 1.200.0 → 1.202.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0e5369f5bb546bc4a434084786180743b3c465f88f079270f4aeb875104a2f5e
-  data.tar.gz: 4b21b756570b018c4f72496c571dbc8dcd58138e3851bab0a50d971df7771782
+  metadata.gz: 20447a1093ec62eddc28f9ba37e3c9e30df3384af011d2d49a49be70a65ec6f6
+  data.tar.gz: 19d9212fdeaaccbf50e1239a7d5ef527d4f02dcda19aa64e1da88bb3d4c5ca89
 SHA512:
-  metadata.gz: bbf4fd1b274d7c25a5d3722e91e30bbcb5817b13a8d32d978a9800a31e4abfc94bc0c1b7e2089380e09cf0deecbbe9e36d4ec7ab397902afd93d64eaf15d5d41
-  data.tar.gz: 528122c7a215f0a6901a12249845b09e8c08fff72f3928ee3800700fd9ea3b168e0af011197f6cd04c5f2094081da87f535af0f73d452202880f0efd1bcf9fe3
+  metadata.gz: bed0f6fa8acb414c51c6ae9a730509b24a5528e22e5b1248e4ad6613826d0a734f3ba0966e0d4c8fd6c0e45707c3b02b48bcf9af7ff10b7db4f174757a9bdc0e
+  data.tar.gz: 3895b91eb914ad77ad5630383dac5d16879587685de544e820de1fecea84e43a8963c832942d898ea5b5ff148a76eed70b7d64bae3934eef7052ef8081f5b62b

data/CHANGELOG.md

@@ -1,6 +1,16 @@
 Unreleased Changes
 ------------------
 
+1.202.0 (2023-08-02)
+------------------
+
+* Feature - SageMaker Inference Recommender introduces a new API GetScalingConfigurationRecommendation to recommend auto scaling policies based on completed Inference Recommender jobs.
+
+1.201.0 (2023-08-01)
+------------------
+
+* Feature - Add Stairs TrafficPattern and FlatInvocations to RecommendationJobStoppingConditions
+
 1.200.0 (2023-07-27)
 ------------------
 

data/VERSION

@@ -1 +1 @@
-1.200.0
+1.202.0

data/lib/aws-sdk-sagemaker/client.rb

@@ -4463,7 +4463,7 @@ module Aws::SageMaker
     #       model_package_version_arn: "ModelPackageArn",
     #       job_duration_in_seconds: 1,
     #       traffic_pattern: {
-    #         traffic_type: "PHASES", # accepts PHASES
+    #         traffic_type: "PHASES", # accepts PHASES, STAIRS
     #         phases: [
     #           {
     #             initial_number_of_users: 1,
@@ -4471,6 +4471,11 @@ module Aws::SageMaker
     #             duration_in_seconds: 1,
     #           },
     #         ],
+    #         stairs: {
+    #           duration_in_seconds: 1,
+    #           number_of_steps: 1,
+    #           users_per_step: 1,
+    #         },
     #       },
     #       resource_limit: {
     #         max_number_of_tests: 1,
@@ -4530,6 +4535,7 @@ module Aws::SageMaker
     #           value_in_milliseconds: 1,
     #         },
     #       ],
+    #       flat_invocations: "Continue", # accepts Continue, Stop
     #     },
     #     output_config: {
     #       kms_key_id: "KmsKeyId",
@@ -12315,11 +12321,14 @@ module Aws::SageMaker
     #   resp.failure_reason #=> String
     #   resp.input_config.model_package_version_arn #=> String
     #   resp.input_config.job_duration_in_seconds #=> Integer
-    #   resp.input_config.traffic_pattern.traffic_type #=> String, one of "PHASES"
+    #   resp.input_config.traffic_pattern.traffic_type #=> String, one of "PHASES", "STAIRS"
     #   resp.input_config.traffic_pattern.phases #=> Array
     #   resp.input_config.traffic_pattern.phases[0].initial_number_of_users #=> Integer
     #   resp.input_config.traffic_pattern.phases[0].spawn_rate #=> Integer
     #   resp.input_config.traffic_pattern.phases[0].duration_in_seconds #=> Integer
+    #   resp.input_config.traffic_pattern.stairs.duration_in_seconds #=> Integer
+    #   resp.input_config.traffic_pattern.stairs.number_of_steps #=> Integer
+    #   resp.input_config.traffic_pattern.stairs.users_per_step #=> Integer
     #   resp.input_config.resource_limit.max_number_of_tests #=> Integer
     #   resp.input_config.resource_limit.max_parallel_of_tests #=> Integer
     #   resp.input_config.endpoint_configurations #=> Array
@@ -12356,6 +12365,7 @@ module Aws::SageMaker
     #   resp.stopping_conditions.model_latency_thresholds #=> Array
     #   resp.stopping_conditions.model_latency_thresholds[0].percentile #=> String
     #   resp.stopping_conditions.model_latency_thresholds[0].value_in_milliseconds #=> Integer
+    #   resp.stopping_conditions.flat_invocations #=> String, one of "Continue", "Stop"
     #   resp.inference_recommendations #=> Array
     #   resp.inference_recommendations[0].metrics.cost_per_hour #=> Float
     #   resp.inference_recommendations[0].metrics.cost_per_inference #=> Float
@@ -14866,6 +14876,90 @@ module Aws::SageMaker
       req.send_request(options)
     end
 
+    # Starts an Amazon SageMaker Inference Recommender autoscaling
+    # recommendation job. Returns recommendations for autoscaling policies
+    # that you can apply to your SageMaker endpoint.
+    #
+    # @option params [required, String] :inference_recommendations_job_name
+    #   The name of a previously completed Inference Recommender job.
+    #
+    # @option params [String] :recommendation_id
+    #   The recommendation ID of a previously completed inference
+    #   recommendation. This ID should come from one of the recommendations
+    #   returned by the job specified in the `InferenceRecommendationsJobName`
+    #   field.
+    #
+    #   Specify either this field or the `EndpointName` field.
+    #
+    # @option params [String] :endpoint_name
+    #   The name of an endpoint benchmarked during a previously completed
+    #   inference recommendation job. This name should come from one of the
+    #   recommendations returned by the job specified in the
+    #   `InferenceRecommendationsJobName` field.
+    #
+    #   Specify either this field or the `RecommendationId` field.
+    #
+    # @option params [Integer] :target_cpu_utilization_per_core
+    #   The percentage of how much utilization you want an instance to use
+    #   before autoscaling. The default value is 50%.
+    #
+    # @option params [Types::ScalingPolicyObjective] :scaling_policy_objective
+    #   An object where you specify the anticipated traffic pattern for an
+    #   endpoint.
+    #
+    # @return [Types::GetScalingConfigurationRecommendationResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
+    #
+    #   * {Types::GetScalingConfigurationRecommendationResponse#inference_recommendations_job_name #inference_recommendations_job_name} => String
+    #   * {Types::GetScalingConfigurationRecommendationResponse#recommendation_id #recommendation_id} => String
+    #   * {Types::GetScalingConfigurationRecommendationResponse#endpoint_name #endpoint_name} => String
+    #   * {Types::GetScalingConfigurationRecommendationResponse#target_cpu_utilization_per_core #target_cpu_utilization_per_core} => Integer
+    #   * {Types::GetScalingConfigurationRecommendationResponse#scaling_policy_objective #scaling_policy_objective} => Types::ScalingPolicyObjective
+    #   * {Types::GetScalingConfigurationRecommendationResponse#metric #metric} => Types::ScalingPolicyMetric
+    #   * {Types::GetScalingConfigurationRecommendationResponse#dynamic_scaling_configuration #dynamic_scaling_configuration} => Types::DynamicScalingConfiguration
+    #
+    # @example Request syntax with placeholder values
+    #
+    #   resp = client.get_scaling_configuration_recommendation({
+    #     inference_recommendations_job_name: "RecommendationJobName", # required
+    #     recommendation_id: "String",
+    #     endpoint_name: "EndpointName",
+    #     target_cpu_utilization_per_core: 1,
+    #     scaling_policy_objective: {
+    #       min_invocations_per_minute: 1,
+    #       max_invocations_per_minute: 1,
+    #     },
+    #   })
+    #
+    # @example Response structure
+    #
+    #   resp.inference_recommendations_job_name #=> String
+    #   resp.recommendation_id #=> String
+    #   resp.endpoint_name #=> String
+    #   resp.target_cpu_utilization_per_core #=> Integer
+    #   resp.scaling_policy_objective.min_invocations_per_minute #=> Integer
+    #   resp.scaling_policy_objective.max_invocations_per_minute #=> Integer
+    #   resp.metric.invocations_per_instance #=> Integer
+    #   resp.metric.model_latency #=> Integer
+    #   resp.dynamic_scaling_configuration.min_capacity #=> Integer
+    #   resp.dynamic_scaling_configuration.max_capacity #=> Integer
+    #   resp.dynamic_scaling_configuration.scale_in_cooldown #=> Integer
+    #   resp.dynamic_scaling_configuration.scale_out_cooldown #=> Integer
+    #   resp.dynamic_scaling_configuration.scaling_policies #=> Array
+    #   resp.dynamic_scaling_configuration.scaling_policies[0].target_tracking.metric_specification.predefined.predefined_metric_type #=> String
+    #   resp.dynamic_scaling_configuration.scaling_policies[0].target_tracking.metric_specification.customized.metric_name #=> String
+    #   resp.dynamic_scaling_configuration.scaling_policies[0].target_tracking.metric_specification.customized.namespace #=> String
+    #   resp.dynamic_scaling_configuration.scaling_policies[0].target_tracking.metric_specification.customized.statistic #=> String, one of "Average", "Minimum", "Maximum", "SampleCount", "Sum"
+    #   resp.dynamic_scaling_configuration.scaling_policies[0].target_tracking.target_value #=> Float
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/GetScalingConfigurationRecommendation AWS API Documentation
+    #
+    # @overload get_scaling_configuration_recommendation(params = {})
+    # @param [Hash] params ({})
+    def get_scaling_configuration_recommendation(params = {}, options = {})
+      req = build_request(:get_scaling_configuration_recommendation, params)
+      req.send_request(options)
+    end
+
     # An auto-complete API for the search functionality in the SageMaker
     # console. It returns suggestions of possible matches for the property
     # name to use in `Search` queries. Provides suggestions for
@@ -23997,7 +24091,7 @@ module Aws::SageMaker
         params: params,
         config: config)
       context[:gem_name] = 'aws-sdk-sagemaker'
-      context[:gem_version] = '1.200.0'
+      context[:gem_version] = '1.202.0'
       Seahorse::Client::Request.new(handlers, context)
     end
 

data/lib/aws-sdk-sagemaker/client_api.rb

@@ -410,6 +410,7 @@ module Aws::SageMaker
     CustomerMetadataKeyList = Shapes::ListShape.new(name: 'CustomerMetadataKeyList')
     CustomerMetadataMap = Shapes::MapShape.new(name: 'CustomerMetadataMap')
     CustomerMetadataValue = Shapes::StringShape.new(name: 'CustomerMetadataValue')
+    CustomizedMetricSpecification = Shapes::StructureShape.new(name: 'CustomizedMetricSpecification')
     DataCaptureConfig = Shapes::StructureShape.new(name: 'DataCaptureConfig')
     DataCaptureConfigSummary = Shapes::StructureShape.new(name: 'DataCaptureConfigSummary')
     DataCatalogConfig = Shapes::StructureShape.new(name: 'DataCatalogConfig')
@@ -667,12 +668,14 @@ module Aws::SageMaker
     DomainSettings = Shapes::StructureShape.new(name: 'DomainSettings')
     DomainSettingsForUpdate = Shapes::StructureShape.new(name: 'DomainSettingsForUpdate')
     DomainStatus = Shapes::StringShape.new(name: 'DomainStatus')
+    Double = Shapes::FloatShape.new(name: 'Double')
     DoubleParameterValue = Shapes::FloatShape.new(name: 'DoubleParameterValue')
     DriftCheckBaselines = Shapes::StructureShape.new(name: 'DriftCheckBaselines')
     DriftCheckBias = Shapes::StructureShape.new(name: 'DriftCheckBias')
     DriftCheckExplainability = Shapes::StructureShape.new(name: 'DriftCheckExplainability')
     DriftCheckModelDataQuality = Shapes::StructureShape.new(name: 'DriftCheckModelDataQuality')
     DriftCheckModelQuality = Shapes::StructureShape.new(name: 'DriftCheckModelQuality')
+    DynamicScalingConfiguration = Shapes::StructureShape.new(name: 'DynamicScalingConfiguration')
     EMRStepMetadata = Shapes::StructureShape.new(name: 'EMRStepMetadata')
     Edge = Shapes::StructureShape.new(name: 'Edge')
     EdgeDeploymentConfig = Shapes::StructureShape.new(name: 'EdgeDeploymentConfig')
@@ -796,6 +799,7 @@ module Aws::SageMaker
     FinalAutoMLJobObjectiveMetric = Shapes::StructureShape.new(name: 'FinalAutoMLJobObjectiveMetric')
     FinalHyperParameterTuningJobObjectiveMetric = Shapes::StructureShape.new(name: 'FinalHyperParameterTuningJobObjectiveMetric')
     FinalMetricDataList = Shapes::ListShape.new(name: 'FinalMetricDataList')
+    FlatInvocations = Shapes::StringShape.new(name: 'FlatInvocations')
     Float = Shapes::FloatShape.new(name: 'Float')
     FlowDefinitionArn = Shapes::StringShape.new(name: 'FlowDefinitionArn')
     FlowDefinitionName = Shapes::StringShape.new(name: 'FlowDefinitionName')
@@ -825,6 +829,8 @@ module Aws::SageMaker
     GetModelPackageGroupPolicyOutput = Shapes::StructureShape.new(name: 'GetModelPackageGroupPolicyOutput')
     GetSagemakerServicecatalogPortfolioStatusInput = Shapes::StructureShape.new(name: 'GetSagemakerServicecatalogPortfolioStatusInput')
     GetSagemakerServicecatalogPortfolioStatusOutput = Shapes::StructureShape.new(name: 'GetSagemakerServicecatalogPortfolioStatusOutput')
+    GetScalingConfigurationRecommendationRequest = Shapes::StructureShape.new(name: 'GetScalingConfigurationRecommendationRequest')
+    GetScalingConfigurationRecommendationResponse = Shapes::StructureShape.new(name: 'GetScalingConfigurationRecommendationResponse')
     GetSearchSuggestionsRequest = Shapes::StructureShape.new(name: 'GetSearchSuggestionsRequest')
     GetSearchSuggestionsResponse = Shapes::StructureShape.new(name: 'GetSearchSuggestionsResponse')
     GitConfig = Shapes::StructureShape.new(name: 'GitConfig')
@@ -1232,6 +1238,7 @@ module Aws::SageMaker
     MetricName = Shapes::StringShape.new(name: 'MetricName')
     MetricRegex = Shapes::StringShape.new(name: 'MetricRegex')
     MetricSetSource = Shapes::StringShape.new(name: 'MetricSetSource')
+    MetricSpecification = Shapes::UnionShape.new(name: 'MetricSpecification')
     MetricValue = Shapes::FloatShape.new(name: 'MetricValue')
     MetricsSource = Shapes::StructureShape.new(name: 'MetricsSource')
     MinimumInstanceMetadataServiceVersion = Shapes::StringShape.new(name: 'MinimumInstanceMetadataServiceVersion')
@@ -1432,6 +1439,7 @@ module Aws::SageMaker
     NotificationConfiguration = Shapes::StructureShape.new(name: 'NotificationConfiguration')
     NotificationTopicArn = Shapes::StringShape.new(name: 'NotificationTopicArn')
     NumberOfHumanWorkersPerDataObject = Shapes::IntegerShape.new(name: 'NumberOfHumanWorkersPerDataObject')
+    NumberOfSteps = Shapes::IntegerShape.new(name: 'NumberOfSteps')
     ObjectiveStatus = Shapes::StringShape.new(name: 'ObjectiveStatus')
     ObjectiveStatusCounter = Shapes::IntegerShape.new(name: 'ObjectiveStatusCounter')
     ObjectiveStatusCounters = Shapes::StructureShape.new(name: 'ObjectiveStatusCounters')
@@ -1502,6 +1510,7 @@ module Aws::SageMaker
     PipelineSummaryList = Shapes::ListShape.new(name: 'PipelineSummaryList')
     PlatformIdentifier = Shapes::StringShape.new(name: 'PlatformIdentifier')
     PolicyString = Shapes::StringShape.new(name: 'PolicyString')
+    PredefinedMetricSpecification = Shapes::StructureShape.new(name: 'PredefinedMetricSpecification')
     PresignedDomainUrl = Shapes::StringShape.new(name: 'PresignedDomainUrl')
     ProbabilityThresholdAttribute = Shapes::FloatShape.new(name: 'ProbabilityThresholdAttribute')
     ProblemType = Shapes::StringShape.new(name: 'ProblemType')
@@ -1701,6 +1710,10 @@ module Aws::SageMaker
     SagemakerServicecatalogStatus = Shapes::StringShape.new(name: 'SagemakerServicecatalogStatus')
     SampleWeightAttributeName = Shapes::StringShape.new(name: 'SampleWeightAttributeName')
     SamplingPercentage = Shapes::IntegerShape.new(name: 'SamplingPercentage')
+    ScalingPolicies = Shapes::ListShape.new(name: 'ScalingPolicies')
+    ScalingPolicy = Shapes::UnionShape.new(name: 'ScalingPolicy')
+    ScalingPolicyMetric = Shapes::StructureShape.new(name: 'ScalingPolicyMetric')
+    ScalingPolicyObjective = Shapes::StructureShape.new(name: 'ScalingPolicyObjective')
     ScheduleConfig = Shapes::StructureShape.new(name: 'ScheduleConfig')
     ScheduleExpression = Shapes::StringShape.new(name: 'ScheduleExpression')
     ScheduleStatus = Shapes::StringShape.new(name: 'ScheduleStatus')
@@ -1770,6 +1783,7 @@ module Aws::SageMaker
     SpawnRate = Shapes::IntegerShape.new(name: 'SpawnRate')
     SplitType = Shapes::StringShape.new(name: 'SplitType')
     StageStatus = Shapes::StringShape.new(name: 'StageStatus')
+    Stairs = Shapes::StructureShape.new(name: 'Stairs')
     StartEdgeDeploymentStageRequest = Shapes::StructureShape.new(name: 'StartEdgeDeploymentStageRequest')
     StartInferenceExperimentRequest = Shapes::StructureShape.new(name: 'StartInferenceExperimentRequest')
     StartInferenceExperimentResponse = Shapes::StructureShape.new(name: 'StartInferenceExperimentResponse')
@@ -1777,6 +1791,7 @@ module Aws::SageMaker
     StartNotebookInstanceInput = Shapes::StructureShape.new(name: 'StartNotebookInstanceInput')
     StartPipelineExecutionRequest = Shapes::StructureShape.new(name: 'StartPipelineExecutionRequest')
     StartPipelineExecutionResponse = Shapes::StructureShape.new(name: 'StartPipelineExecutionResponse')
+    Statistic = Shapes::StringShape.new(name: 'Statistic')
     StatusDetails = Shapes::StringShape.new(name: 'StatusDetails')
     StatusMessage = Shapes::StringShape.new(name: 'StatusMessage')
     StepDescription = Shapes::StringShape.new(name: 'StepDescription')
@@ -1841,6 +1856,7 @@ module Aws::SageMaker
     TargetPlatformAccelerator = Shapes::StringShape.new(name: 'TargetPlatformAccelerator')
     TargetPlatformArch = Shapes::StringShape.new(name: 'TargetPlatformArch')
     TargetPlatformOs = Shapes::StringShape.new(name: 'TargetPlatformOs')
+    TargetTrackingScalingPolicyConfiguration = Shapes::StructureShape.new(name: 'TargetTrackingScalingPolicyConfiguration')
     TaskAvailabilityLifetimeInSeconds = Shapes::IntegerShape.new(name: 'TaskAvailabilityLifetimeInSeconds')
     TaskCount = Shapes::IntegerShape.new(name: 'TaskCount')
     TaskDescription = Shapes::StringShape.new(name: 'TaskDescription')
@@ -2027,7 +2043,9 @@ module Aws::SageMaker
     UserProfileSortKey = Shapes::StringShape.new(name: 'UserProfileSortKey')
     UserProfileStatus = Shapes::StringShape.new(name: 'UserProfileStatus')
     UserSettings = Shapes::StructureShape.new(name: 'UserSettings')
+    UsersPerStep = Shapes::IntegerShape.new(name: 'UsersPerStep')
     UtilizationMetric = Shapes::FloatShape.new(name: 'UtilizationMetric')
+    UtilizationPercentagePerCore = Shapes::IntegerShape.new(name: 'UtilizationPercentagePerCore')
     ValidationFraction = Shapes::FloatShape.new(name: 'ValidationFraction')
     VariantName = Shapes::StringShape.new(name: 'VariantName')
     VariantProperty = Shapes::StructureShape.new(name: 'VariantProperty')
@@ -3396,6 +3414,11 @@ module Aws::SageMaker
     CustomerMetadataMap.key = Shapes::ShapeRef.new(shape: CustomerMetadataKey)
     CustomerMetadataMap.value = Shapes::ShapeRef.new(shape: CustomerMetadataValue)
 
+    CustomizedMetricSpecification.add_member(:metric_name, Shapes::ShapeRef.new(shape: String, location_name: "MetricName"))
+    CustomizedMetricSpecification.add_member(:namespace, Shapes::ShapeRef.new(shape: String, location_name: "Namespace"))
+    CustomizedMetricSpecification.add_member(:statistic, Shapes::ShapeRef.new(shape: Statistic, location_name: "Statistic"))
+    CustomizedMetricSpecification.struct_class = Types::CustomizedMetricSpecification
+
     DataCaptureConfig.add_member(:enable_capture, Shapes::ShapeRef.new(shape: EnableCapture, location_name: "EnableCapture"))
     DataCaptureConfig.add_member(:initial_sampling_percentage, Shapes::ShapeRef.new(shape: SamplingPercentage, required: true, location_name: "InitialSamplingPercentage"))
     DataCaptureConfig.add_member(:destination_s3_uri, Shapes::ShapeRef.new(shape: DestinationS3Uri, required: true, location_name: "DestinationS3Uri"))
@@ -4854,6 +4877,13 @@ module Aws::SageMaker
     DriftCheckModelQuality.add_member(:constraints, Shapes::ShapeRef.new(shape: MetricsSource, location_name: "Constraints"))
     DriftCheckModelQuality.struct_class = Types::DriftCheckModelQuality
 
+    DynamicScalingConfiguration.add_member(:min_capacity, Shapes::ShapeRef.new(shape: Integer, location_name: "MinCapacity"))
+    DynamicScalingConfiguration.add_member(:max_capacity, Shapes::ShapeRef.new(shape: Integer, location_name: "MaxCapacity"))
+    DynamicScalingConfiguration.add_member(:scale_in_cooldown, Shapes::ShapeRef.new(shape: Integer, location_name: "ScaleInCooldown"))
+    DynamicScalingConfiguration.add_member(:scale_out_cooldown, Shapes::ShapeRef.new(shape: Integer, location_name: "ScaleOutCooldown"))
+    DynamicScalingConfiguration.add_member(:scaling_policies, Shapes::ShapeRef.new(shape: ScalingPolicies, location_name: "ScalingPolicies"))
+    DynamicScalingConfiguration.struct_class = Types::DynamicScalingConfiguration
+
     EMRStepMetadata.add_member(:cluster_id, Shapes::ShapeRef.new(shape: String256, location_name: "ClusterId"))
     EMRStepMetadata.add_member(:step_id, Shapes::ShapeRef.new(shape: String256, location_name: "StepId"))
     EMRStepMetadata.add_member(:step_name, Shapes::ShapeRef.new(shape: String256, location_name: "StepName"))
@@ -5221,6 +5251,22 @@ module Aws::SageMaker
     GetSagemakerServicecatalogPortfolioStatusOutput.add_member(:status, Shapes::ShapeRef.new(shape: SagemakerServicecatalogStatus, location_name: "Status"))
     GetSagemakerServicecatalogPortfolioStatusOutput.struct_class = Types::GetSagemakerServicecatalogPortfolioStatusOutput
 
+    GetScalingConfigurationRecommendationRequest.add_member(:inference_recommendations_job_name, Shapes::ShapeRef.new(shape: RecommendationJobName, required: true, location_name: "InferenceRecommendationsJobName"))
+    GetScalingConfigurationRecommendationRequest.add_member(:recommendation_id, Shapes::ShapeRef.new(shape: String, location_name: "RecommendationId"))
+    GetScalingConfigurationRecommendationRequest.add_member(:endpoint_name, Shapes::ShapeRef.new(shape: EndpointName, location_name: "EndpointName"))
+    GetScalingConfigurationRecommendationRequest.add_member(:target_cpu_utilization_per_core, Shapes::ShapeRef.new(shape: UtilizationPercentagePerCore, location_name: "TargetCpuUtilizationPerCore"))
+    GetScalingConfigurationRecommendationRequest.add_member(:scaling_policy_objective, Shapes::ShapeRef.new(shape: ScalingPolicyObjective, location_name: "ScalingPolicyObjective"))
+    GetScalingConfigurationRecommendationRequest.struct_class = Types::GetScalingConfigurationRecommendationRequest
+
+    GetScalingConfigurationRecommendationResponse.add_member(:inference_recommendations_job_name, Shapes::ShapeRef.new(shape: RecommendationJobName, location_name: "InferenceRecommendationsJobName"))
+    GetScalingConfigurationRecommendationResponse.add_member(:recommendation_id, Shapes::ShapeRef.new(shape: String, location_name: "RecommendationId"))
+    GetScalingConfigurationRecommendationResponse.add_member(:endpoint_name, Shapes::ShapeRef.new(shape: EndpointName, location_name: "EndpointName"))
+    GetScalingConfigurationRecommendationResponse.add_member(:target_cpu_utilization_per_core, Shapes::ShapeRef.new(shape: UtilizationPercentagePerCore, location_name: "TargetCpuUtilizationPerCore"))
+    GetScalingConfigurationRecommendationResponse.add_member(:scaling_policy_objective, Shapes::ShapeRef.new(shape: ScalingPolicyObjective, location_name: "ScalingPolicyObjective"))
+    GetScalingConfigurationRecommendationResponse.add_member(:metric, Shapes::ShapeRef.new(shape: ScalingPolicyMetric, location_name: "Metric"))
+    GetScalingConfigurationRecommendationResponse.add_member(:dynamic_scaling_configuration, Shapes::ShapeRef.new(shape: DynamicScalingConfiguration, location_name: "DynamicScalingConfiguration"))
+    GetScalingConfigurationRecommendationResponse.struct_class = Types::GetScalingConfigurationRecommendationResponse
+
     GetSearchSuggestionsRequest.add_member(:resource, Shapes::ShapeRef.new(shape: ResourceType, required: true, location_name: "Resource"))
     GetSearchSuggestionsRequest.add_member(:suggestion_query, Shapes::ShapeRef.new(shape: SuggestionQuery, location_name: "SuggestionQuery"))
     GetSearchSuggestionsRequest.struct_class = Types::GetSearchSuggestionsRequest
@@ -6782,6 +6828,14 @@ module Aws::SageMaker
 
     MetricDefinitionList.member = Shapes::ShapeRef.new(shape: MetricDefinition)
 
+    MetricSpecification.add_member(:predefined, Shapes::ShapeRef.new(shape: PredefinedMetricSpecification, location_name: "Predefined"))
+    MetricSpecification.add_member(:customized, Shapes::ShapeRef.new(shape: CustomizedMetricSpecification, location_name: "Customized"))
+    MetricSpecification.add_member(:unknown, Shapes::ShapeRef.new(shape: nil, location_name: 'unknown'))
+    MetricSpecification.add_member_subclass(:predefined, Types::MetricSpecification::Predefined)
+    MetricSpecification.add_member_subclass(:customized, Types::MetricSpecification::Customized)
+    MetricSpecification.add_member_subclass(:unknown, Types::MetricSpecification::Unknown)
+    MetricSpecification.struct_class = Types::MetricSpecification
+
     MetricsSource.add_member(:content_type, Shapes::ShapeRef.new(shape: ContentType, required: true, location_name: "ContentType"))
     MetricsSource.add_member(:content_digest, Shapes::ShapeRef.new(shape: ContentDigest, location_name: "ContentDigest"))
     MetricsSource.add_member(:s3_uri, Shapes::ShapeRef.new(shape: S3Uri, required: true, location_name: "S3Uri"))
@@ -7586,6 +7640,9 @@ module Aws::SageMaker
 
     PipelineSummaryList.member = Shapes::ShapeRef.new(shape: PipelineSummary)
 
+    PredefinedMetricSpecification.add_member(:predefined_metric_type, Shapes::ShapeRef.new(shape: String, location_name: "PredefinedMetricType"))
+    PredefinedMetricSpecification.struct_class = Types::PredefinedMetricSpecification
+
     ProcessingClusterConfig.add_member(:instance_count, Shapes::ShapeRef.new(shape: ProcessingInstanceCount, required: true, location_name: "InstanceCount"))
     ProcessingClusterConfig.add_member(:instance_type, Shapes::ShapeRef.new(shape: ProcessingInstanceType, required: true, location_name: "InstanceType"))
     ProcessingClusterConfig.add_member(:volume_size_in_gb, Shapes::ShapeRef.new(shape: ProcessingVolumeSizeInGB, required: true, location_name: "VolumeSizeInGB"))
@@ -7936,6 +7993,7 @@ module Aws::SageMaker
 
     RecommendationJobStoppingConditions.add_member(:max_invocations, Shapes::ShapeRef.new(shape: Integer, location_name: "MaxInvocations"))
     RecommendationJobStoppingConditions.add_member(:model_latency_thresholds, Shapes::ShapeRef.new(shape: ModelLatencyThresholds, location_name: "ModelLatencyThresholds"))
+    RecommendationJobStoppingConditions.add_member(:flat_invocations, Shapes::ShapeRef.new(shape: FlatInvocations, location_name: "FlatInvocations"))
     RecommendationJobStoppingConditions.struct_class = Types::RecommendationJobStoppingConditions
 
     RecommendationJobSupportedContentTypes.member = Shapes::ShapeRef.new(shape: String)
@@ -8088,6 +8146,22 @@ module Aws::SageMaker
 
     SageMakerImageVersionAliases.member = Shapes::ShapeRef.new(shape: SageMakerImageVersionAlias)
 
+    ScalingPolicies.member = Shapes::ShapeRef.new(shape: ScalingPolicy)
+
+    ScalingPolicy.add_member(:target_tracking, Shapes::ShapeRef.new(shape: TargetTrackingScalingPolicyConfiguration, location_name: "TargetTracking"))
+    ScalingPolicy.add_member(:unknown, Shapes::ShapeRef.new(shape: nil, location_name: 'unknown'))
+    ScalingPolicy.add_member_subclass(:target_tracking, Types::ScalingPolicy::TargetTracking)
+    ScalingPolicy.add_member_subclass(:unknown, Types::ScalingPolicy::Unknown)
+    ScalingPolicy.struct_class = Types::ScalingPolicy
+
+    ScalingPolicyMetric.add_member(:invocations_per_instance, Shapes::ShapeRef.new(shape: Integer, location_name: "InvocationsPerInstance"))
+    ScalingPolicyMetric.add_member(:model_latency, Shapes::ShapeRef.new(shape: Integer, location_name: "ModelLatency"))
+    ScalingPolicyMetric.struct_class = Types::ScalingPolicyMetric
+
+    ScalingPolicyObjective.add_member(:min_invocations_per_minute, Shapes::ShapeRef.new(shape: Integer, location_name: "MinInvocationsPerMinute"))
+    ScalingPolicyObjective.add_member(:max_invocations_per_minute, Shapes::ShapeRef.new(shape: Integer, location_name: "MaxInvocationsPerMinute"))
+    ScalingPolicyObjective.struct_class = Types::ScalingPolicyObjective
+
     ScheduleConfig.add_member(:schedule_expression, Shapes::ShapeRef.new(shape: ScheduleExpression, required: true, location_name: "ScheduleExpression"))
     ScheduleConfig.struct_class = Types::ScheduleConfig
 
@@ -8226,6 +8300,11 @@ module Aws::SageMaker
     SpaceSettings.add_member(:kernel_gateway_app_settings, Shapes::ShapeRef.new(shape: KernelGatewayAppSettings, location_name: "KernelGatewayAppSettings"))
     SpaceSettings.struct_class = Types::SpaceSettings
 
+    Stairs.add_member(:duration_in_seconds, Shapes::ShapeRef.new(shape: TrafficDurationInSeconds, location_name: "DurationInSeconds"))
+    Stairs.add_member(:number_of_steps, Shapes::ShapeRef.new(shape: NumberOfSteps, location_name: "NumberOfSteps"))
+    Stairs.add_member(:users_per_step, Shapes::ShapeRef.new(shape: UsersPerStep, location_name: "UsersPerStep"))
+    Stairs.struct_class = Types::Stairs
+
     StartEdgeDeploymentStageRequest.add_member(:edge_deployment_plan_name, Shapes::ShapeRef.new(shape: EntityName, required: true, location_name: "EdgeDeploymentPlanName"))
     StartEdgeDeploymentStageRequest.add_member(:stage_name, Shapes::ShapeRef.new(shape: EntityName, required: true, location_name: "StageName"))
     StartEdgeDeploymentStageRequest.struct_class = Types::StartEdgeDeploymentStageRequest
@@ -8361,6 +8440,10 @@ module Aws::SageMaker
     TargetPlatform.add_member(:accelerator, Shapes::ShapeRef.new(shape: TargetPlatformAccelerator, location_name: "Accelerator"))
     TargetPlatform.struct_class = Types::TargetPlatform
 
+    TargetTrackingScalingPolicyConfiguration.add_member(:metric_specification, Shapes::ShapeRef.new(shape: MetricSpecification, location_name: "MetricSpecification"))
+    TargetTrackingScalingPolicyConfiguration.add_member(:target_value, Shapes::ShapeRef.new(shape: Double, location_name: "TargetValue"))
+    TargetTrackingScalingPolicyConfiguration.struct_class = Types::TargetTrackingScalingPolicyConfiguration
+
     TaskKeywords.member = Shapes::ShapeRef.new(shape: TaskKeyword)
 
     TensorBoardAppSettings.add_member(:default_resource_spec, Shapes::ShapeRef.new(shape: ResourceSpec, location_name: "DefaultResourceSpec"))
@@ -8400,6 +8483,7 @@ module Aws::SageMaker
 
     TrafficPattern.add_member(:traffic_type, Shapes::ShapeRef.new(shape: TrafficType, location_name: "TrafficType"))
     TrafficPattern.add_member(:phases, Shapes::ShapeRef.new(shape: Phases, location_name: "Phases"))
+    TrafficPattern.add_member(:stairs, Shapes::ShapeRef.new(shape: Stairs, location_name: "Stairs"))
     TrafficPattern.struct_class = Types::TrafficPattern
 
     TrafficRoutingConfig.add_member(:type, Shapes::ShapeRef.new(shape: TrafficRoutingConfigType, required: true, location_name: "Type"))
@@ -10686,6 +10770,15 @@ module Aws::SageMaker
         o.output = Shapes::ShapeRef.new(shape: GetSagemakerServicecatalogPortfolioStatusOutput)
       end)
 
+      api.add_operation(:get_scaling_configuration_recommendation, Seahorse::Model::Operation.new.tap do |o|
+        o.name = "GetScalingConfigurationRecommendation"
+        o.http_method = "POST"
+        o.http_request_uri = "/"
+        o.input = Shapes::ShapeRef.new(shape: GetScalingConfigurationRecommendationRequest)
+        o.output = Shapes::ShapeRef.new(shape: GetScalingConfigurationRecommendationResponse)
+        o.errors << Shapes::ShapeRef.new(shape: ResourceNotFound)
+      end)
+
       api.add_operation(:get_search_suggestions, Seahorse::Model::Operation.new.tap do |o|
         o.name = "GetSearchSuggestions"
         o.http_method = "POST"

data/lib/aws-sdk-sagemaker/endpoints.rb

@@ -2378,6 +2378,20 @@ module Aws::SageMaker
       end
     end
 
+    class GetScalingConfigurationRecommendation
+      def self.build(context)
+        unless context.config.regional_endpoint
+          endpoint = context.config.endpoint.to_s
+        end
+        Aws::SageMaker::EndpointParameters.new(
+          region: context.config.region,
+          use_dual_stack: context.config.use_dualstack_endpoint,
+          use_fips: context.config.use_fips_endpoint,
+          endpoint: endpoint,
+        )
+      end
+    end
+
     class GetSearchSuggestions
       def self.build(context)
         unless context.config.regional_endpoint

data/lib/aws-sdk-sagemaker/plugins/endpoints.rb

@@ -394,6 +394,8 @@ module Aws::SageMaker
             Aws::SageMaker::Endpoints::GetModelPackageGroupPolicy.build(context)
           when :get_sagemaker_servicecatalog_portfolio_status
             Aws::SageMaker::Endpoints::GetSagemakerServicecatalogPortfolioStatus.build(context)
+          when :get_scaling_configuration_recommendation
+            Aws::SageMaker::Endpoints::GetScalingConfigurationRecommendation.build(context)
           when :get_search_suggestions
             Aws::SageMaker::Endpoints::GetSearchSuggestions.build(context)
           when :import_hub_content

data/lib/aws-sdk-sagemaker/types.rb

@@ -2858,7 +2858,7 @@ module Aws::SageMaker
     #   @return [String]
     #
     # @!attribute [rw] end_time_offset
-    #   If specified, monitoring jobs substract this time from the end time.
+    #   If specified, monitoring jobs subtract this time from the end time.
     #   For information about using offsets for scheduling monitoring jobs,
     #   see [Schedule Model Quality Monitoring Jobs][1].
     #
@@ -8837,6 +8837,30 @@ module Aws::SageMaker
       include Aws::Structure
     end
 
+    # A customized metric.
+    #
+    # @!attribute [rw] metric_name
+    #   The name of the customized metric.
+    #   @return [String]
+    #
+    # @!attribute [rw] namespace
+    #   The namespace of the customized metric.
+    #   @return [String]
+    #
+    # @!attribute [rw] statistic
+    #   The statistic of the customized metric.
+    #   @return [String]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/CustomizedMetricSpecification AWS API Documentation
+    #
+    class CustomizedMetricSpecification < Struct.new(
+      :metric_name,
+      :namespace,
+      :statistic)
+      SENSITIVE = []
+      include Aws::Structure
+    end
+
     # Configuration to control how SageMaker captures inference data.
     #
     # @!attribute [rw] enable_capture
@@ -11956,6 +11980,12 @@ module Aws::SageMaker
     #     for information about the failure. [DeleteEndpoint][4] is the only
     #     operation that can be performed on a failed endpoint.
     #
+    #   * `UpdateRollbackFailed`: Both the rolling deployment and
+    #     auto-rollback failed. Your endpoint is in service with a mix of
+    #     the old and new endpoint configurations. For information about how
+    #     to remedy this issue and restore the endpoint's status to
+    #     `InService`, see [Rolling Deployments][6].
+    #
     #
     #
     #   [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateEndpoint.html
@@ -11963,6 +11993,7 @@ module Aws::SageMaker
     #   [3]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_UpdateEndpointWeightsAndCapacities.html
     #   [4]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DeleteEndpoint.html
     #   [5]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeEndpoint.html
+    #   [6]: https://docs.aws.amazon.com/sagemaker/latest/dg/deployment-guardrails-rolling.html
     #   @return [String]
     #
     # @!attribute [rw] failure_reason
@@ -16488,6 +16519,43 @@ module Aws::SageMaker
       include Aws::Structure
     end
 
+    # An object with the recommended values for you to specify when creating
+    # an autoscaling policy.
+    #
+    # @!attribute [rw] min_capacity
+    #   The recommended minimum capacity to specify for your autoscaling
+    #   policy.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] max_capacity
+    #   The recommended maximum capacity to specify for your autoscaling
+    #   policy.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] scale_in_cooldown
+    #   The recommended scale in cooldown time for your autoscaling policy.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] scale_out_cooldown
+    #   The recommended scale out cooldown time for your autoscaling policy.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] scaling_policies
+    #   An object of the scaling policies for each metric.
+    #   @return [Array<Types::ScalingPolicy>]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DynamicScalingConfiguration AWS API Documentation
+    #
+    class DynamicScalingConfiguration < Struct.new(
+      :min_capacity,
+      :max_capacity,
+      :scale_in_cooldown,
+      :scale_out_cooldown,
+      :scaling_policies)
+      SENSITIVE = []
+      include Aws::Structure
+    end
+
     # The configurations and outcomes of an Amazon EMR step execution.
     #
     # @!attribute [rw] cluster_id
@@ -18380,6 +18448,99 @@ module Aws::SageMaker
       include Aws::Structure
     end
 
+    # @!attribute [rw] inference_recommendations_job_name
+    #   The name of a previously completed Inference Recommender job.
+    #   @return [String]
+    #
+    # @!attribute [rw] recommendation_id
+    #   The recommendation ID of a previously completed inference
+    #   recommendation. This ID should come from one of the recommendations
+    #   returned by the job specified in the
+    #   `InferenceRecommendationsJobName` field.
+    #
+    #   Specify either this field or the `EndpointName` field.
+    #   @return [String]
+    #
+    # @!attribute [rw] endpoint_name
+    #   The name of an endpoint benchmarked during a previously completed
+    #   inference recommendation job. This name should come from one of the
+    #   recommendations returned by the job specified in the
+    #   `InferenceRecommendationsJobName` field.
+    #
+    #   Specify either this field or the `RecommendationId` field.
+    #   @return [String]
+    #
+    # @!attribute [rw] target_cpu_utilization_per_core
+    #   The percentage of how much utilization you want an instance to use
+    #   before autoscaling. The default value is 50%.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] scaling_policy_objective
+    #   An object where you specify the anticipated traffic pattern for an
+    #   endpoint.
+    #   @return [Types::ScalingPolicyObjective]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/GetScalingConfigurationRecommendationRequest AWS API Documentation
+    #
+    class GetScalingConfigurationRecommendationRequest < Struct.new(
+      :inference_recommendations_job_name,
+      :recommendation_id,
+      :endpoint_name,
+      :target_cpu_utilization_per_core,
+      :scaling_policy_objective)
+      SENSITIVE = []
+      include Aws::Structure
+    end
+
+    # @!attribute [rw] inference_recommendations_job_name
+    #   The name of a previously completed Inference Recommender job.
+    #   @return [String]
+    #
+    # @!attribute [rw] recommendation_id
+    #   The recommendation ID of a previously completed inference
+    #   recommendation.
+    #   @return [String]
+    #
+    # @!attribute [rw] endpoint_name
+    #   The name of an endpoint benchmarked during a previously completed
+    #   Inference Recommender job.
+    #   @return [String]
+    #
+    # @!attribute [rw] target_cpu_utilization_per_core
+    #   The percentage of how much utilization you want an instance to use
+    #   before autoscaling, which you specified in the request. The default
+    #   value is 50%.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] scaling_policy_objective
+    #   An object representing the anticipated traffic pattern for an
+    #   endpoint that you specified in the request.
+    #   @return [Types::ScalingPolicyObjective]
+    #
+    # @!attribute [rw] metric
+    #   An object with a list of metrics that were benchmarked during the
+    #   previously completed Inference Recommender job.
+    #   @return [Types::ScalingPolicyMetric]
+    #
+    # @!attribute [rw] dynamic_scaling_configuration
+    #   An object with the recommended values for you to specify when
+    #   creating an autoscaling policy.
+    #   @return [Types::DynamicScalingConfiguration]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/GetScalingConfigurationRecommendationResponse AWS API Documentation
+    #
+    class GetScalingConfigurationRecommendationResponse < Struct.new(
+      :inference_recommendations_job_name,
+      :recommendation_id,
+      :endpoint_name,
+      :target_cpu_utilization_per_core,
+      :scaling_policy_objective,
+      :metric,
+      :dynamic_scaling_configuration)
+      SENSITIVE = []
+      include Aws::Structure
+    end
+
     # @!attribute [rw] resource
     #   The name of the SageMaker resource to search for.
     #   @return [String]
@@ -27839,6 +28000,33 @@ module Aws::SageMaker
       include Aws::Structure
     end
 
+    # An object containing information about a metric.
+    #
+    # @note MetricSpecification is a union - when returned from an API call exactly one value will be set and the returned type will be a subclass of MetricSpecification corresponding to the set member.
+    #
+    # @!attribute [rw] predefined
+    #   Information about a predefined metric.
+    #   @return [Types::PredefinedMetricSpecification]
+    #
+    # @!attribute [rw] customized
+    #   Information about a customized metric.
+    #   @return [Types::CustomizedMetricSpecification]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/MetricSpecification AWS API Documentation
+    #
+    class MetricSpecification < Struct.new(
+      :predefined,
+      :customized,
+      :unknown)
+      SENSITIVE = []
+      include Aws::Structure
+      include Aws::Structure::Union
+
+      class Predefined < MetricSpecification; end
+      class Customized < MetricSpecification; end
+      class Unknown < MetricSpecification; end
+    end
+
     # Details about the metrics source.
     #
     # @!attribute [rw] content_type
@@ -28827,7 +29015,8 @@ module Aws::SageMaker
     # The model latency threshold.
     #
     # @!attribute [rw] percentile
-    #   The model latency percentile threshold.
+    #   The model latency percentile threshold. For custom load tests,
+    #   specify the value as `P95`.
     #   @return [String]
     #
     # @!attribute [rw] value_in_milliseconds
@@ -31739,7 +31928,8 @@ module Aws::SageMaker
     # Defines the traffic pattern.
     #
     # @!attribute [rw] initial_number_of_users
-    #   Specifies how many concurrent users to start with.
+    #   Specifies how many concurrent users to start with. The value should
+    #   be between 1 and 3.
     #   @return [Integer]
     #
     # @!attribute [rw] spawn_rate
@@ -31747,7 +31937,9 @@ module Aws::SageMaker
     #   @return [Integer]
     #
     # @!attribute [rw] duration_in_seconds
-    #   Specifies how long traffic phase should be.
+    #   Specifies how long a traffic phase should be. For custom load tests,
+    #   the value should be between 120 and 3600. This value should not
+    #   exceed `JobDurationInSeconds`.
     #   @return [Integer]
     #
     # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/Phase AWS API Documentation
@@ -32257,6 +32449,21 @@ module Aws::SageMaker
       include Aws::Structure
     end
 
+    # A specification for a predefined metric.
+    #
+    # @!attribute [rw] predefined_metric_type
+    #   The metric type. You can only apply SageMaker metric types to
+    #   SageMaker endpoints.
+    #   @return [String]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/PredefinedMetricSpecification AWS API Documentation
+    #
+    class PredefinedMetricSpecification < Struct.new(
+      :predefined_metric_type)
+      SENSITIVE = []
+      include Aws::Structure
+    end
+
     # Configuration for the cluster used to run a processing job.
     #
     # @!attribute [rw] instance_count
@@ -34263,7 +34470,8 @@ module Aws::SageMaker
     #   @return [String]
     #
     # @!attribute [rw] job_duration_in_seconds
-    #   Specifies the maximum duration of the job, in seconds.&gt;
+    #   Specifies the maximum duration of the job, in seconds. The maximum
+    #   value is 7200.
     #   @return [Integer]
     #
     # @!attribute [rw] traffic_pattern
@@ -34462,11 +34670,19 @@ module Aws::SageMaker
     #   container.
     #   @return [Array<Types::ModelLatencyThreshold>]
     #
+    # @!attribute [rw] flat_invocations
+    #   Stops a load test when the number of invocations (TPS) peaks and
+    #   flattens, which means that the instance has reached capacity. The
+    #   default value is `Stop`. If you want the load test to continue after
+    #   invocations have flattened, set the value to `Continue`.
+    #   @return [String]
+    #
     # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/RecommendationJobStoppingConditions AWS API Documentation
     #
     class RecommendationJobStoppingConditions < Struct.new(
       :max_invocations,
-      :model_latency_thresholds)
+      :model_latency_thresholds,
+      :flat_invocations)
       SENSITIVE = []
       include Aws::Structure
     end
@@ -35450,6 +35666,75 @@ module Aws::SageMaker
       include Aws::Structure
     end
 
+    # An object containing a recommended scaling policy.
+    #
+    # @note ScalingPolicy is a union - when returned from an API call exactly one value will be set and the returned type will be a subclass of ScalingPolicy corresponding to the set member.
+    #
+    # @!attribute [rw] target_tracking
+    #   A target tracking scaling policy. Includes support for predefined or
+    #   customized metrics.
+    #   @return [Types::TargetTrackingScalingPolicyConfiguration]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ScalingPolicy AWS API Documentation
+    #
+    class ScalingPolicy < Struct.new(
+      :target_tracking,
+      :unknown)
+      SENSITIVE = []
+      include Aws::Structure
+      include Aws::Structure::Union
+
+      class TargetTracking < ScalingPolicy; end
+      class Unknown < ScalingPolicy; end
+    end
+
+    # The metric for a scaling policy.
+    #
+    # @!attribute [rw] invocations_per_instance
+    #   The number of invocations sent to a model, normalized by
+    #   `InstanceCount` in each ProductionVariant. `1/numberOfInstances` is
+    #   sent as the value on each request, where `numberOfInstances` is the
+    #   number of active instances for the ProductionVariant behind the
+    #   endpoint at the time of the request.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] model_latency
+    #   The interval of time taken by a model to respond as viewed from
+    #   SageMaker. This interval includes the local communication times
+    #   taken to send the request and to fetch the response from the
+    #   container of a model and the time taken to complete the inference in
+    #   the container.
+    #   @return [Integer]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ScalingPolicyMetric AWS API Documentation
+    #
+    class ScalingPolicyMetric < Struct.new(
+      :invocations_per_instance,
+      :model_latency)
+      SENSITIVE = []
+      include Aws::Structure
+    end
+
+    # An object where you specify the anticipated traffic pattern for an
+    # endpoint.
+    #
+    # @!attribute [rw] min_invocations_per_minute
+    #   The minimum number of expected requests to your endpoint per minute.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] max_invocations_per_minute
+    #   The maximum number of expected requests to your endpoint per minute.
+    #   @return [Integer]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ScalingPolicyObjective AWS API Documentation
+    #
+    class ScalingPolicyObjective < Struct.new(
+      :min_invocations_per_minute,
+      :max_invocations_per_minute)
+      SENSITIVE = []
+      include Aws::Structure
+    end
+
     # Configuration details about the monitoring schedule.
     #
     # @!attribute [rw] schedule_expression
@@ -36338,6 +36623,34 @@ module Aws::SageMaker
       include Aws::Structure
     end
 
+    # Defines the stairs traffic pattern for an Inference Recommender load
+    # test. This pattern type consists of multiple steps where the number of
+    # users increases at each step.
+    #
+    # Specify either the stairs or phases traffic pattern.
+    #
+    # @!attribute [rw] duration_in_seconds
+    #   Defines how long each traffic step should be.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] number_of_steps
+    #   Specifies how many steps to perform during traffic.
+    #   @return [Integer]
+    #
+    # @!attribute [rw] users_per_step
+    #   Specifies how many new users to spawn in each step.
+    #   @return [Integer]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/Stairs AWS API Documentation
+    #
+    class Stairs < Struct.new(
+      :duration_in_seconds,
+      :number_of_steps,
+      :users_per_step)
+      SENSITIVE = []
+      include Aws::Structure
+    end
+
     # @!attribute [rw] edge_deployment_plan_name
     #   The name of the edge deployment plan to start.
     #   @return [String]
@@ -37111,6 +37424,35 @@ module Aws::SageMaker
       include Aws::Structure
     end
 
+    # A target tracking scaling policy. Includes support for predefined or
+    # customized metrics.
+    #
+    # When using the [PutScalingPolicy][1] API, this parameter is required
+    # when you are creating a policy with the policy type
+    # `TargetTrackingScaling`.
+    #
+    #
+    #
+    # [1]: https://docs.aws.amazon.com/autoscaling/application/APIReference/API_PutScalingPolicy.html
+    #
+    # @!attribute [rw] metric_specification
+    #   An object containing information about a metric.
+    #   @return [Types::MetricSpecification]
+    #
+    # @!attribute [rw] target_value
+    #   The recommended target value to specify for the metric when creating
+    #   a scaling policy.
+    #   @return [Float]
+    #
+    # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/TargetTrackingScalingPolicyConfiguration AWS API Documentation
+    #
+    class TargetTrackingScalingPolicyConfiguration < Struct.new(
+      :metric_specification,
+      :target_value)
+      SENSITIVE = []
+      include Aws::Structure
+    end
+
     # The TensorBoard app settings.
     #
     # @!attribute [rw] default_resource_spec
@@ -37402,18 +37744,23 @@ module Aws::SageMaker
     # Defines the traffic pattern of the load test.
     #
     # @!attribute [rw] traffic_type
-    #   Defines the traffic patterns.
+    #   Defines the traffic patterns. Choose either `PHASES` or `STAIRS`.
     #   @return [String]
     #
     # @!attribute [rw] phases
     #   Defines the phases traffic specification.
     #   @return [Array<Types::Phase>]
     #
+    # @!attribute [rw] stairs
+    #   Defines the stairs traffic pattern.
+    #   @return [Types::Stairs]
+    #
     # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/TrafficPattern AWS API Documentation
     #
     class TrafficPattern < Struct.new(
       :traffic_type,
-      :phases)
+      :phases,
+      :stairs)
       SENSITIVE = []
       include Aws::Structure
     end

data/lib/aws-sdk-sagemaker.rb

@@ -53,6 +53,6 @@ require_relative 'aws-sdk-sagemaker/customizations'
 # @!group service
 module Aws::SageMaker
 
-  GEM_VERSION = '1.200.0'
+  GEM_VERSION = '1.202.0'
 
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: aws-sdk-sagemaker
 version: !ruby/object:Gem::Version
-  version: 1.200.0
+  version: 1.202.0
 platform: ruby
 authors:
 - Amazon Web Services
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-07-27 00:00:00.000000000 Z
+date: 2023-08-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: aws-sdk-core