aws-sdk-sagemaker 1.414.0 → 1.416.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -2934,7 +2934,7 @@ module Aws::SageMaker
2934
2934
  # },
2935
2935
  # image_id: "ImageId",
2936
2936
  # auto_patch_config: {
2937
- # patching_strategy: "WhenIdle", # required, accepts WhenIdle, WhenAllIdle
2937
+ # patching_strategy: "WhenIdle", # required, accepts WhenIdle, WhenAllIdle, DrainFirst
2938
2938
  # patch_schedule: {
2939
2939
  # next_patch_date: Time.now,
2940
2940
  # },
@@ -2956,6 +2956,11 @@ module Aws::SageMaker
2956
2956
  # },
2957
2957
  # ],
2958
2958
  # },
2959
+ # idle_wait_window_in_seconds: 1,
2960
+ # drain_policy: {
2961
+ # drain_timeout_in_seconds: 1,
2962
+ # on_drain_timeout: "Force", # accepts Force, Skip
2963
+ # },
2959
2964
  # },
2960
2965
  # image_release_version: "ImageReleaseVersion",
2961
2966
  # kubernetes_config: {
@@ -4942,6 +4947,18 @@ module Aws::SageMaker
4942
4947
  # @option params [Types::MetricsConfig] :metrics_config
4943
4948
  # The configuration parameters for utilization metrics.
4944
4949
  #
4950
+ # @option params [Types::CrossRegionTrafficConfig] :cross_region_traffic_config
4951
+ # The cross-Region traffic cap for the endpoint. The cap limits the
4952
+ # requests per second that the endpoint accepts from inference routers
4953
+ # in other Amazon Web Services Regions. If you don't specify it, these
4954
+ # requests aren't limited. You can't specify it with serverless
4955
+ # variants or with `AsyncInferenceConfig`. For information about the
4956
+ # cross-Region traffic cap, see [Limit requests from other Regions][1].
4957
+ #
4958
+ #
4959
+ #
4960
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/inference-router.html#inference-router-cross-region-limit
4961
+ #
4945
4962
  # @return [Types::CreateEndpointConfigOutput] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
4946
4963
  #
4947
4964
  # * {Types::CreateEndpointConfigOutput#endpoint_config_arn #endpoint_config_arn} => String
@@ -4990,7 +5007,7 @@ module Aws::SageMaker
4990
5007
  # },
4991
5008
  # },
4992
5009
  # routing_config: {
4993
- # routing_strategy: "LEAST_OUTSTANDING_REQUESTS", # required, accepts LEAST_OUTSTANDING_REQUESTS, RANDOM, PREFIX_AWARE
5010
+ # routing_strategy: "LEAST_OUTSTANDING_REQUESTS", # required, accepts LEAST_OUTSTANDING_REQUESTS, RANDOM, PREFIX_AWARE, KV_AWARE
4994
5011
  # prefix_aware_routing_config: {
4995
5012
  # prefix_length: 1,
4996
5013
  # concurrency_threshold: 1,
@@ -5117,7 +5134,7 @@ module Aws::SageMaker
5117
5134
  # },
5118
5135
  # },
5119
5136
  # routing_config: {
5120
- # routing_strategy: "LEAST_OUTSTANDING_REQUESTS", # required, accepts LEAST_OUTSTANDING_REQUESTS, RANDOM, PREFIX_AWARE
5137
+ # routing_strategy: "LEAST_OUTSTANDING_REQUESTS", # required, accepts LEAST_OUTSTANDING_REQUESTS, RANDOM, PREFIX_AWARE, KV_AWARE
5121
5138
  # prefix_aware_routing_config: {
5122
5139
  # prefix_length: 1,
5123
5140
  # concurrency_threshold: 1,
@@ -5146,6 +5163,10 @@ module Aws::SageMaker
5146
5163
  # enable_detailed_observability: false,
5147
5164
  # metric_publish_frequency_in_seconds: 1,
5148
5165
  # },
5166
+ # cross_region_traffic_config: {
5167
+ # max_invocations_per_second_per_instance: 1, # required
5168
+ # max_cross_region_percent_per_endpoint: 1, # required
5169
+ # },
5149
5170
  # })
5150
5171
  #
5151
5172
  # @example Response structure
@@ -7043,17 +7064,149 @@ module Aws::SageMaker
7043
7064
  req.send_request(options)
7044
7065
  end
7045
7066
 
7067
+ # Creates an inference router. An inference router is a Regional
7068
+ # SageMaker AI resource that sends each inference request to one of the
7069
+ # endpoints or inference components that you list for a routing target.
7070
+ # These destinations can be in any Amazon Web Services account, and in
7071
+ # any Amazon Web Services Region in the same partition as the router.
7072
+ #
7073
+ # The router's status is `Creating` until it's ready to use. SageMaker
7074
+ # AI checks that each endpoint is a real-time or asynchronous endpoint,
7075
+ # as the routing target requires, and rejects one of the wrong type. It
7076
+ # checks endpoints in other accounts with your credentials, so the
7077
+ # endpoint policy must allow `sagemaker:DescribeEndpoint`. SageMaker AI
7078
+ # accepts an endpoint that it can't check, for example one that
7079
+ # doesn't exist. Requests to it fail if it's the wrong type or
7080
+ # doesn't exist. To see the status of each destination, call
7081
+ # [DescribeInferenceRouter][1] and see the `RoutingStatus` of each
7082
+ # destination.
7083
+ #
7084
+ # To invoke the router, use `InvokeInferenceRouter` or
7085
+ # `InvokeInferenceRouterAsync`. For information about inference routers,
7086
+ # see [Route requests across Regions with inference routers][2].
7087
+ #
7088
+ #
7089
+ #
7090
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeInferenceRouter.html
7091
+ # [2]: https://docs.aws.amazon.com/sagemaker/latest/dg/inference-router.html
7092
+ #
7093
+ # @option params [required, String] :inference_router_name
7094
+ # The name of the inference router. The name must be unique within your
7095
+ # account in the current Amazon Web Services Region. The name is case
7096
+ # sensitive.
7097
+ #
7098
+ # @option params [required, Array<Types::RoutingTarget>] :routing_targets
7099
+ # The routing targets of the inference router. Each routing target is a
7100
+ # model that callers name when they invoke the router, and the
7101
+ # destinations that serve it.
7102
+ #
7103
+ # @option params [String] :client_token
7104
+ # A unique, case-sensitive identifier that you provide to ensure the
7105
+ # idempotency of the request. If you reuse a client token, use the same
7106
+ # request parameters. Otherwise, the request fails.
7107
+ #
7108
+ # **A suitable default value is auto-generated.** You should normally
7109
+ # not need to pass this option.**
7110
+ #
7111
+ # @option params [Array<Types::Tag>] :tags
7112
+ # A list of key-value pairs to apply to the inference router. For
7113
+ # information about tagging, see [Tagging Amazon Web Services
7114
+ # resources][1] in the *Amazon Web Services General Reference*.
7115
+ #
7116
+ #
7117
+ #
7118
+ # [1]: https://docs.aws.amazon.com/general/latest/gr/aws_tagging.html
7119
+ #
7120
+ # @return [Types::CreateInferenceRouterOutput] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
7121
+ #
7122
+ # * {Types::CreateInferenceRouterOutput#inference_router_arn #inference_router_arn} => String
7123
+ #
7124
+ # @example Request syntax with placeholder values
7125
+ #
7126
+ # resp = client.create_inference_router({
7127
+ # inference_router_name: "InferenceRouterName", # required
7128
+ # routing_targets: [ # required
7129
+ # {
7130
+ # name: "RoutingTargetName", # required
7131
+ # destinations: [ # required
7132
+ # {
7133
+ # resource_arn: "DestinationResourceArn", # required
7134
+ # routing_status: "Routable", # accepts Routable, Unroutable, Unknown
7135
+ # },
7136
+ # ],
7137
+ # async_inference_config: {
7138
+ # output_config: { # required
7139
+ # s3_output_path: "InferenceRouterS3Uri", # required
7140
+ # s3_failure_path: "InferenceRouterS3Uri", # required
7141
+ # kms_key_id: "KmsKeyId",
7142
+ # notification_config: {
7143
+ # success_topic: "InferenceRouterSnsTopicArn",
7144
+ # error_topic: "InferenceRouterSnsTopicArn",
7145
+ # },
7146
+ # },
7147
+ # client_config: {
7148
+ # max_time_in_backlog_seconds: 1,
7149
+ # },
7150
+ # },
7151
+ # routing_preference: "LocalRegionFirst", # accepts LocalRegionFirst, InstancePoolOrder
7152
+ # },
7153
+ # ],
7154
+ # client_token: "InferenceRouterClientToken",
7155
+ # tags: [
7156
+ # {
7157
+ # key: "TagKey", # required
7158
+ # value: "TagValue", # required
7159
+ # },
7160
+ # ],
7161
+ # })
7162
+ #
7163
+ # @example Response structure
7164
+ #
7165
+ # resp.inference_router_arn #=> String
7166
+ #
7167
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/CreateInferenceRouter AWS API Documentation
7168
+ #
7169
+ # @overload create_inference_router(params = {})
7170
+ # @param [Hash] params ({})
7171
+ def create_inference_router(params = {}, options = {})
7172
+ req = build_request(:create_inference_router, params)
7173
+ req.send_request(options)
7174
+ end
7175
+
7046
7176
  # Creates a model customization job in Amazon SageMaker. A job runs a
7047
7177
  # workload based on the job category and configuration you provide. You
7048
7178
  # specify the job category, a schema-versioned configuration document,
7049
7179
  # and an IAM role that grants Amazon SageMaker permission to access
7050
7180
  # resources on your behalf.
7051
7181
  #
7052
- # Use the `AgentRFT` category to fine-tune a model using multi-turn
7053
- # reinforcement learning with reward signals. Use the
7054
- # `AgentRFTEvaluation` category to evaluate a fine-tuned or base model
7055
- # by running multi-turn rollouts against a held-out prompt dataset and
7056
- # computing metrics such as pass@k and mean reward.
7182
+ # Specify one of the following job categories, which determines the type
7183
+ # of workload that the job runs:
7184
+ #
7185
+ # AgentRFT
7186
+ #
7187
+ # : Fine-tune a model using multi-turn reinforcement learning with
7188
+ # reward signals.
7189
+ #
7190
+ # AgentRFTEvaluation
7191
+ #
7192
+ # : Evaluate a fine-tuned or base model by running multi-turn rollouts
7193
+ # against a held-out prompt dataset. This category computes metrics
7194
+ # such as pass@k and mean reward.
7195
+ #
7196
+ # DataQualityEvaluation
7197
+ #
7198
+ # : Evaluate the quality of a generated or bring-your-own dataset using
7199
+ # quality metrics or your own specification.
7200
+ #
7201
+ # DataTransformation
7202
+ #
7203
+ # : Transform datasets into the format you need, for example for model
7204
+ # customization.
7205
+ #
7206
+ # SyntheticDataGeneration
7207
+ #
7208
+ # : Generate synthetic training and testing datasets from your context
7209
+ # data and specifications.
7057
7210
  #
7058
7211
  # Before creating a job, call `ListJobSchemaVersions` and
7059
7212
  # `DescribeJobSchemaVersion` to retrieve the configuration schema for
@@ -7114,7 +7267,7 @@ module Aws::SageMaker
7114
7267
  # resp = client.create_job({
7115
7268
  # job_name: "JobName", # required
7116
7269
  # role_arn: "RoleArn", # required
7117
- # job_category: "AgentRFT", # required, accepts AgentRFT, AgentRFTEvaluation
7270
+ # job_category: "AgentRFT", # required, accepts AgentRFT, SyntheticDataGeneration, DataQualityEvaluation, AgentRFTEvaluation, DataTransformation
7118
7271
  # job_config_schema_version: "JobSchemaVersion", # required
7119
7272
  # job_config_document: "JobConfigDocument", # required
7120
7273
  # tags: [
@@ -13179,6 +13332,42 @@ module Aws::SageMaker
13179
13332
  req.send_request(options)
13180
13333
  end
13181
13334
 
13335
+ # Deletes the resource-based policy that is attached to an endpoint.
13336
+ # After you delete the policy, other accounts can no longer send
13337
+ # requests to the endpoint through an inference router. The endpoint
13338
+ # continues to serve requests from your account. If the endpoint
13339
+ # doesn't have a policy, the operation succeeds.
13340
+ #
13341
+ # The change typically takes effect within 30 seconds, and can take up
13342
+ # to 5 minutes. To revoke access for one account only, use
13343
+ # [PutEndpointPolicy][1] to replace the policy with one that doesn't
13344
+ # include that account.
13345
+ #
13346
+ #
13347
+ #
13348
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_PutEndpointPolicy.html
13349
+ #
13350
+ # @option params [required, String] :endpoint_name
13351
+ # The name of the endpoint. The endpoint must be in your account and in
13352
+ # the current Amazon Web Services Region.
13353
+ #
13354
+ # @return [Struct] Returns an empty {Seahorse::Client::Response response}.
13355
+ #
13356
+ # @example Request syntax with placeholder values
13357
+ #
13358
+ # resp = client.delete_endpoint_policy({
13359
+ # endpoint_name: "EndpointName", # required
13360
+ # })
13361
+ #
13362
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DeleteEndpointPolicy AWS API Documentation
13363
+ #
13364
+ # @overload delete_endpoint_policy(params = {})
13365
+ # @param [Hash] params ({})
13366
+ def delete_endpoint_policy(params = {}, options = {})
13367
+ req = build_request(:delete_endpoint_policy, params)
13368
+ req.send_request(options)
13369
+ end
13370
+
13182
13371
  # Deletes an SageMaker experiment. All trials associated with the
13183
13372
  # experiment must be deleted first. Use the [ListTrials][1] API to get a
13184
13373
  # list of the trials associated with the experiment.
@@ -13524,6 +13713,32 @@ module Aws::SageMaker
13524
13713
  req.send_request(options)
13525
13714
  end
13526
13715
 
13716
+ # Deletes an inference router. The endpoints and inference components
13717
+ # that the router sends requests to aren't deleted.
13718
+ #
13719
+ # If the router is being created, updated, or deleted, you get a
13720
+ # `ConflictException`.
13721
+ #
13722
+ # @option params [required, String] :inference_router_name
13723
+ # The name of the inference router to delete.
13724
+ #
13725
+ # @return [Struct] Returns an empty {Seahorse::Client::Response response}.
13726
+ #
13727
+ # @example Request syntax with placeholder values
13728
+ #
13729
+ # resp = client.delete_inference_router({
13730
+ # inference_router_name: "InferenceRouterName", # required
13731
+ # })
13732
+ #
13733
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DeleteInferenceRouter AWS API Documentation
13734
+ #
13735
+ # @overload delete_inference_router(params = {})
13736
+ # @param [Hash] params ({})
13737
+ def delete_inference_router(params = {}, options = {})
13738
+ req = build_request(:delete_inference_router, params)
13739
+ req.send_request(options)
13740
+ end
13741
+
13527
13742
  # Deletes a job. This operation is idempotent. If the job is currently
13528
13743
  # running, you must stop it before deleting it by calling `StopJob`.
13529
13744
  #
@@ -13547,7 +13762,7 @@ module Aws::SageMaker
13547
13762
  #
13548
13763
  # resp = client.delete_job({
13549
13764
  # job_name: "JobName", # required
13550
- # job_category: "AgentRFT", # required, accepts AgentRFT, AgentRFTEvaluation
13765
+ # job_category: "AgentRFT", # required, accepts AgentRFT, SyntheticDataGeneration, DataQualityEvaluation, AgentRFTEvaluation, DataTransformation
13551
13766
  # })
13552
13767
  #
13553
13768
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DeleteJob AWS API Documentation
@@ -15482,7 +15697,7 @@ module Aws::SageMaker
15482
15697
  # resp.instance_groups[0].scheduled_update_config.deployment_config.wait_interval_in_seconds #=> Integer
15483
15698
  # resp.instance_groups[0].scheduled_update_config.deployment_config.auto_rollback_configuration #=> Array
15484
15699
  # resp.instance_groups[0].scheduled_update_config.deployment_config.auto_rollback_configuration[0].alarm_name #=> String
15485
- # resp.instance_groups[0].auto_patch_config.patching_strategy #=> String, one of "WhenIdle", "WhenAllIdle"
15700
+ # resp.instance_groups[0].auto_patch_config.patching_strategy #=> String, one of "WhenIdle", "WhenAllIdle", "DrainFirst"
15486
15701
  # resp.instance_groups[0].auto_patch_config.current_patch_schedule.next_patch_date #=> Time
15487
15702
  # resp.instance_groups[0].auto_patch_config.desired_patch_schedule.next_patch_date #=> Time
15488
15703
  # resp.instance_groups[0].auto_patch_config.deployment_config.rolling_update_policy.maximum_batch_size.type #=> String, one of "INSTANCE_COUNT", "CAPACITY_PERCENTAGE"
@@ -15492,6 +15707,9 @@ module Aws::SageMaker
15492
15707
  # resp.instance_groups[0].auto_patch_config.deployment_config.wait_interval_in_seconds #=> Integer
15493
15708
  # resp.instance_groups[0].auto_patch_config.deployment_config.auto_rollback_configuration #=> Array
15494
15709
  # resp.instance_groups[0].auto_patch_config.deployment_config.auto_rollback_configuration[0].alarm_name #=> String
15710
+ # resp.instance_groups[0].auto_patch_config.idle_wait_window_in_seconds #=> Integer
15711
+ # resp.instance_groups[0].auto_patch_config.drain_policy.drain_timeout_in_seconds #=> Integer
15712
+ # resp.instance_groups[0].auto_patch_config.drain_policy.on_drain_timeout #=> String, one of "Force", "Skip"
15495
15713
  # resp.instance_groups[0].current_image_id #=> String
15496
15714
  # resp.instance_groups[0].desired_image_id #=> String
15497
15715
  # resp.instance_groups[0].current_image_release_version #=> String
@@ -15663,6 +15881,7 @@ module Aws::SageMaker
15663
15881
  # resp.event_details.event_details.event_metadata.slurm_health.reason #=> String, one of "DaemonDown", "DaemonDisabled", "DbUnreachable"
15664
15882
  # resp.event_details.description #=> String
15665
15883
  # resp.event_details.event_level #=> String, one of "Info", "Warn", "Error"
15884
+ # resp.event_details.operation_id #=> String
15666
15885
  #
15667
15886
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeClusterEvent AWS API Documentation
15668
15887
  #
@@ -15676,7 +15895,9 @@ module Aws::SageMaker
15676
15895
  # Retrieves information about an extension installed on a SageMaker
15677
15896
  # HyperPod cluster. The information includes the extension's version,
15678
15897
  # configuration, and status, and a summary of how many cluster nodes are
15679
- # running the current version and configuration of the extension.
15898
+ # running the current version and configuration of the extension. If the
15899
+ # extension has problems on some nodes, the response also lists the
15900
+ # problems and how many nodes each one affects.
15680
15901
  #
15681
15902
  # @option params [required, String] :cluster_arn
15682
15903
  # The Amazon Resource Name (ARN) of the SageMaker HyperPod cluster on
@@ -15692,6 +15913,7 @@ module Aws::SageMaker
15692
15913
  #
15693
15914
  # * {Types::DescribeClusterExtensionResponse#extension_details #extension_details} => Types::ExtensionDetails
15694
15915
  # * {Types::DescribeClusterExtensionResponse#node_summary #node_summary} => Types::ExtensionNodeSummary
15916
+ # * {Types::DescribeClusterExtensionResponse#health #health} => Types::ExtensionHealth
15695
15917
  #
15696
15918
  # @example Request syntax with placeholder values
15697
15919
  #
@@ -15712,6 +15934,10 @@ module Aws::SageMaker
15712
15934
  # resp.extension_details.last_modified_time #=> Time
15713
15935
  # resp.node_summary.total_count #=> Integer
15714
15936
  # resp.node_summary.active_count #=> Integer
15937
+ # resp.health.issues #=> Array
15938
+ # resp.health.issues[0].code #=> String, one of "InvalidConfiguration", "ConfigurationConflict", "AccessDenied", "ResourceNotFound", "DependencyUnreachable", "ExtensionUnhealthy", "InternalFailure", "Other"
15939
+ # resp.health.issues[0].message #=> String
15940
+ # resp.health.issues[0].node_count #=> Integer
15715
15941
  #
15716
15942
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeClusterExtension AWS API Documentation
15717
15943
  #
@@ -15813,6 +16039,68 @@ module Aws::SageMaker
15813
16039
  req.send_request(options)
15814
16040
  end
15815
16041
 
16042
+ # Retrieves detailed information about a specific operation for a given
16043
+ # HyperPod cluster, such as a scaling operation or a deep health check.
16044
+ # Use the `OperationId` returned by the `ListClusterOperations`
16045
+ # operation to identify the operation to describe.
16046
+ #
16047
+ # @option params [required, String] :cluster_name
16048
+ # The name or Amazon Resource Name (ARN) of the HyperPod cluster
16049
+ # associated with the operation.
16050
+ #
16051
+ # @option params [required, String] :operation_id
16052
+ # The unique identifier (UUID) of the operation to describe. This ID can
16053
+ # be obtained from the `ListClusterOperations` operation.
16054
+ #
16055
+ # @return [Types::DescribeClusterOperationResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
16056
+ #
16057
+ # * {Types::DescribeClusterOperationResponse#operation_id #operation_id} => String
16058
+ # * {Types::DescribeClusterOperationResponse#cluster_arn #cluster_arn} => String
16059
+ # * {Types::DescribeClusterOperationResponse#type #type} => String
16060
+ # * {Types::DescribeClusterOperationResponse#status #status} => String
16061
+ # * {Types::DescribeClusterOperationResponse#creation_time #creation_time} => Time
16062
+ # * {Types::DescribeClusterOperationResponse#scaling_details #scaling_details} => Array&lt;Types::ClusterOperationScalingDetails&gt;
16063
+ # * {Types::DescribeClusterOperationResponse#deep_health_check_details #deep_health_check_details} => Array&lt;Types::InstanceGroupHealthCheckConfiguration&gt;
16064
+ #
16065
+ # @example Request syntax with placeholder values
16066
+ #
16067
+ # resp = client.describe_cluster_operation({
16068
+ # cluster_name: "ClusterNameOrArn", # required
16069
+ # operation_id: "ClusterOperationId", # required
16070
+ # })
16071
+ #
16072
+ # @example Response structure
16073
+ #
16074
+ # resp.operation_id #=> String
16075
+ # resp.cluster_arn #=> String
16076
+ # resp.type #=> String, one of "Scaling", "DeepHealthCheck"
16077
+ # resp.status #=> String, one of "InProgress", "Superseded", "Done"
16078
+ # resp.creation_time #=> Time
16079
+ # resp.scaling_details #=> Array
16080
+ # resp.scaling_details[0].instance_group_name #=> String
16081
+ # resp.scaling_details[0].status #=> String, one of "InProgress", "Superseded", "Done"
16082
+ # resp.scaling_details[0].original.current_count #=> Integer
16083
+ # resp.scaling_details[0].original.target_count #=> Integer
16084
+ # resp.scaling_details[0].original.min_count #=> Integer
16085
+ # resp.scaling_details[0].desired.current_count #=> Integer
16086
+ # resp.scaling_details[0].desired.target_count #=> Integer
16087
+ # resp.scaling_details[0].desired.min_count #=> Integer
16088
+ # resp.deep_health_check_details #=> Array
16089
+ # resp.deep_health_check_details[0].instance_group_name #=> String
16090
+ # resp.deep_health_check_details[0].instance_ids #=> Array
16091
+ # resp.deep_health_check_details[0].instance_ids[0] #=> String
16092
+ # resp.deep_health_check_details[0].deep_health_checks #=> Array
16093
+ # resp.deep_health_check_details[0].deep_health_checks[0] #=> String, one of "InstanceStress", "InstanceConnectivity", "MultiNodeTrainingValidation"
16094
+ #
16095
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeClusterOperation AWS API Documentation
16096
+ #
16097
+ # @overload describe_cluster_operation(params = {})
16098
+ # @param [Hash] params ({})
16099
+ def describe_cluster_operation(params = {}, options = {})
16100
+ req = build_request(:describe_cluster_operation, params)
16101
+ req.send_request(options)
16102
+ end
16103
+
15816
16104
  # Description of the cluster policy. This policy is used for task
15817
16105
  # prioritization and fair-share allocation. This helps prioritize
15818
16106
  # critical workloads and distributes idle compute across entities.
@@ -16800,6 +17088,7 @@ module Aws::SageMaker
16800
17088
  # * {Types::DescribeEndpointOutput#explainer_config #explainer_config} => Types::ExplainerConfig
16801
17089
  # * {Types::DescribeEndpointOutput#shadow_production_variants #shadow_production_variants} => Array&lt;Types::ProductionVariantSummary&gt;
16802
17090
  # * {Types::DescribeEndpointOutput#metrics_config #metrics_config} => Types::MetricsConfig
17091
+ # * {Types::DescribeEndpointOutput#cross_region_traffic_config #cross_region_traffic_config} => Types::CrossRegionTrafficConfig
16803
17092
  #
16804
17093
  # @example Request syntax with placeholder values
16805
17094
  #
@@ -16841,7 +17130,7 @@ module Aws::SageMaker
16841
17130
  # resp.production_variants[0].managed_instance_scaling.scale_in_policy.strategy #=> String, one of "IDLE_RELEASE", "CONSOLIDATION"
16842
17131
  # resp.production_variants[0].managed_instance_scaling.scale_in_policy.maximum_step_size #=> Integer
16843
17132
  # resp.production_variants[0].managed_instance_scaling.scale_in_policy.cooldown_in_minutes #=> Integer
16844
- # resp.production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE"
17133
+ # resp.production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE", "KV_AWARE"
16845
17134
  # resp.production_variants[0].routing_config.prefix_aware_routing_config.prefix_length #=> Integer
16846
17135
  # resp.production_variants[0].routing_config.prefix_aware_routing_config.concurrency_threshold #=> Integer
16847
17136
  # resp.production_variants[0].capacity_reservation_config.ml_reservation_arn #=> String
@@ -16922,7 +17211,7 @@ module Aws::SageMaker
16922
17211
  # resp.pending_deployment_summary.production_variants[0].managed_instance_scaling.scale_in_policy.strategy #=> String, one of "IDLE_RELEASE", "CONSOLIDATION"
16923
17212
  # resp.pending_deployment_summary.production_variants[0].managed_instance_scaling.scale_in_policy.maximum_step_size #=> Integer
16924
17213
  # resp.pending_deployment_summary.production_variants[0].managed_instance_scaling.scale_in_policy.cooldown_in_minutes #=> Integer
16925
- # resp.pending_deployment_summary.production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE"
17214
+ # resp.pending_deployment_summary.production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE", "KV_AWARE"
16926
17215
  # resp.pending_deployment_summary.production_variants[0].routing_config.prefix_aware_routing_config.prefix_length #=> Integer
16927
17216
  # resp.pending_deployment_summary.production_variants[0].routing_config.prefix_aware_routing_config.concurrency_threshold #=> Integer
16928
17217
  # resp.pending_deployment_summary.production_variants[0].accelerator_partition_layouts #=> Array
@@ -16960,7 +17249,7 @@ module Aws::SageMaker
16960
17249
  # resp.pending_deployment_summary.shadow_production_variants[0].managed_instance_scaling.scale_in_policy.strategy #=> String, one of "IDLE_RELEASE", "CONSOLIDATION"
16961
17250
  # resp.pending_deployment_summary.shadow_production_variants[0].managed_instance_scaling.scale_in_policy.maximum_step_size #=> Integer
16962
17251
  # resp.pending_deployment_summary.shadow_production_variants[0].managed_instance_scaling.scale_in_policy.cooldown_in_minutes #=> Integer
16963
- # resp.pending_deployment_summary.shadow_production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE"
17252
+ # resp.pending_deployment_summary.shadow_production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE", "KV_AWARE"
16964
17253
  # resp.pending_deployment_summary.shadow_production_variants[0].routing_config.prefix_aware_routing_config.prefix_length #=> Integer
16965
17254
  # resp.pending_deployment_summary.shadow_production_variants[0].routing_config.prefix_aware_routing_config.concurrency_threshold #=> Integer
16966
17255
  # resp.pending_deployment_summary.shadow_production_variants[0].accelerator_partition_layouts #=> Array
@@ -17018,7 +17307,7 @@ module Aws::SageMaker
17018
17307
  # resp.shadow_production_variants[0].managed_instance_scaling.scale_in_policy.strategy #=> String, one of "IDLE_RELEASE", "CONSOLIDATION"
17019
17308
  # resp.shadow_production_variants[0].managed_instance_scaling.scale_in_policy.maximum_step_size #=> Integer
17020
17309
  # resp.shadow_production_variants[0].managed_instance_scaling.scale_in_policy.cooldown_in_minutes #=> Integer
17021
- # resp.shadow_production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE"
17310
+ # resp.shadow_production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE", "KV_AWARE"
17022
17311
  # resp.shadow_production_variants[0].routing_config.prefix_aware_routing_config.prefix_length #=> Integer
17023
17312
  # resp.shadow_production_variants[0].routing_config.prefix_aware_routing_config.concurrency_threshold #=> Integer
17024
17313
  # resp.shadow_production_variants[0].capacity_reservation_config.ml_reservation_arn #=> String
@@ -17037,6 +17326,8 @@ module Aws::SageMaker
17037
17326
  # resp.metrics_config.enable_enhanced_metrics #=> Boolean
17038
17327
  # resp.metrics_config.enable_detailed_observability #=> Boolean
17039
17328
  # resp.metrics_config.metric_publish_frequency_in_seconds #=> Integer
17329
+ # resp.cross_region_traffic_config.max_invocations_per_second_per_instance #=> Integer
17330
+ # resp.cross_region_traffic_config.max_cross_region_percent_per_endpoint #=> Integer
17040
17331
  #
17041
17332
  #
17042
17333
  # The following waiters are defined for this operation (see {Client#wait_until} for detailed usage):
@@ -17074,6 +17365,7 @@ module Aws::SageMaker
17074
17365
  # * {Types::DescribeEndpointConfigOutput#vpc_config #vpc_config} => Types::VpcConfig
17075
17366
  # * {Types::DescribeEndpointConfigOutput#enable_network_isolation #enable_network_isolation} => Boolean
17076
17367
  # * {Types::DescribeEndpointConfigOutput#metrics_config #metrics_config} => Types::MetricsConfig
17368
+ # * {Types::DescribeEndpointConfigOutput#cross_region_traffic_config #cross_region_traffic_config} => Types::CrossRegionTrafficConfig
17077
17369
  #
17078
17370
  # @example Request syntax with placeholder values
17079
17371
  #
@@ -17112,7 +17404,7 @@ module Aws::SageMaker
17112
17404
  # resp.production_variants[0].managed_instance_scaling.scale_in_policy.strategy #=> String, one of "IDLE_RELEASE", "CONSOLIDATION"
17113
17405
  # resp.production_variants[0].managed_instance_scaling.scale_in_policy.maximum_step_size #=> Integer
17114
17406
  # resp.production_variants[0].managed_instance_scaling.scale_in_policy.cooldown_in_minutes #=> Integer
17115
- # resp.production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE"
17407
+ # resp.production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE", "KV_AWARE"
17116
17408
  # resp.production_variants[0].routing_config.prefix_aware_routing_config.prefix_length #=> Integer
17117
17409
  # resp.production_variants[0].routing_config.prefix_aware_routing_config.concurrency_threshold #=> Integer
17118
17410
  # resp.production_variants[0].inference_ami_version #=> String, one of "al2-ami-sagemaker-inference-gpu-2", "al2-ami-sagemaker-inference-gpu-2-1", "al2-ami-sagemaker-inference-gpu-3-1", "al2-ami-sagemaker-inference-neuron-2", "al2023-ami-sagemaker-inference-gpu-4-1"
@@ -17191,7 +17483,7 @@ module Aws::SageMaker
17191
17483
  # resp.shadow_production_variants[0].managed_instance_scaling.scale_in_policy.strategy #=> String, one of "IDLE_RELEASE", "CONSOLIDATION"
17192
17484
  # resp.shadow_production_variants[0].managed_instance_scaling.scale_in_policy.maximum_step_size #=> Integer
17193
17485
  # resp.shadow_production_variants[0].managed_instance_scaling.scale_in_policy.cooldown_in_minutes #=> Integer
17194
- # resp.shadow_production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE"
17486
+ # resp.shadow_production_variants[0].routing_config.routing_strategy #=> String, one of "LEAST_OUTSTANDING_REQUESTS", "RANDOM", "PREFIX_AWARE", "KV_AWARE"
17195
17487
  # resp.shadow_production_variants[0].routing_config.prefix_aware_routing_config.prefix_length #=> Integer
17196
17488
  # resp.shadow_production_variants[0].routing_config.prefix_aware_routing_config.concurrency_threshold #=> Integer
17197
17489
  # resp.shadow_production_variants[0].inference_ami_version #=> String, one of "al2-ami-sagemaker-inference-gpu-2", "al2-ami-sagemaker-inference-gpu-2-1", "al2-ami-sagemaker-inference-gpu-3-1", "al2-ami-sagemaker-inference-neuron-2", "al2023-ami-sagemaker-inference-gpu-4-1"
@@ -17209,6 +17501,8 @@ module Aws::SageMaker
17209
17501
  # resp.metrics_config.enable_enhanced_metrics #=> Boolean
17210
17502
  # resp.metrics_config.enable_detailed_observability #=> Boolean
17211
17503
  # resp.metrics_config.metric_publish_frequency_in_seconds #=> Integer
17504
+ # resp.cross_region_traffic_config.max_invocations_per_second_per_instance #=> Integer
17505
+ # resp.cross_region_traffic_config.max_cross_region_percent_per_endpoint #=> Integer
17212
17506
  #
17213
17507
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeEndpointConfig AWS API Documentation
17214
17508
  #
@@ -18427,6 +18721,67 @@ module Aws::SageMaker
18427
18721
  req.send_request(options)
18428
18722
  end
18429
18723
 
18724
+ # Returns information about an inference router, including the routing
18725
+ # status of each destination.
18726
+ #
18727
+ # SageMaker AI checks each destination with your permissions when you
18728
+ # call this operation, so callers with different permissions can see a
18729
+ # different `RoutingStatus`. To get the router's tags, use
18730
+ # [ListTags][1].
18731
+ #
18732
+ #
18733
+ #
18734
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_ListTags.html
18735
+ #
18736
+ # @option params [required, String] :inference_router_name
18737
+ # The name of the inference router to describe.
18738
+ #
18739
+ # @return [Types::DescribeInferenceRouterOutput] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
18740
+ #
18741
+ # * {Types::DescribeInferenceRouterOutput#inference_router_name #inference_router_name} => String
18742
+ # * {Types::DescribeInferenceRouterOutput#inference_router_arn #inference_router_arn} => String
18743
+ # * {Types::DescribeInferenceRouterOutput#routing_targets #routing_targets} => Array&lt;Types::RoutingTarget&gt;
18744
+ # * {Types::DescribeInferenceRouterOutput#status #status} => String
18745
+ # * {Types::DescribeInferenceRouterOutput#status_reason #status_reason} => String
18746
+ # * {Types::DescribeInferenceRouterOutput#creation_time #creation_time} => Time
18747
+ # * {Types::DescribeInferenceRouterOutput#last_modified_time #last_modified_time} => Time
18748
+ #
18749
+ # @example Request syntax with placeholder values
18750
+ #
18751
+ # resp = client.describe_inference_router({
18752
+ # inference_router_name: "InferenceRouterName", # required
18753
+ # })
18754
+ #
18755
+ # @example Response structure
18756
+ #
18757
+ # resp.inference_router_name #=> String
18758
+ # resp.inference_router_arn #=> String
18759
+ # resp.routing_targets #=> Array
18760
+ # resp.routing_targets[0].name #=> String
18761
+ # resp.routing_targets[0].destinations #=> Array
18762
+ # resp.routing_targets[0].destinations[0].resource_arn #=> String
18763
+ # resp.routing_targets[0].destinations[0].routing_status #=> String, one of "Routable", "Unroutable", "Unknown"
18764
+ # resp.routing_targets[0].async_inference_config.output_config.s3_output_path #=> String
18765
+ # resp.routing_targets[0].async_inference_config.output_config.s3_failure_path #=> String
18766
+ # resp.routing_targets[0].async_inference_config.output_config.kms_key_id #=> String
18767
+ # resp.routing_targets[0].async_inference_config.output_config.notification_config.success_topic #=> String
18768
+ # resp.routing_targets[0].async_inference_config.output_config.notification_config.error_topic #=> String
18769
+ # resp.routing_targets[0].async_inference_config.client_config.max_time_in_backlog_seconds #=> Integer
18770
+ # resp.routing_targets[0].routing_preference #=> String, one of "LocalRegionFirst", "InstancePoolOrder"
18771
+ # resp.status #=> String, one of "InService", "Creating", "Updating", "Failed", "Deleting"
18772
+ # resp.status_reason #=> String
18773
+ # resp.creation_time #=> Time
18774
+ # resp.last_modified_time #=> Time
18775
+ #
18776
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeInferenceRouter AWS API Documentation
18777
+ #
18778
+ # @overload describe_inference_router(params = {})
18779
+ # @param [Hash] params ({})
18780
+ def describe_inference_router(params = {}, options = {})
18781
+ req = build_request(:describe_inference_router, params)
18782
+ req.send_request(options)
18783
+ end
18784
+
18430
18785
  # Returns detailed information about a job, including its current
18431
18786
  # status, secondary status, configuration, and timestamps. Use
18432
18787
  # `SecondaryStatus` for granular progress tracking and
@@ -18470,7 +18825,7 @@ module Aws::SageMaker
18470
18825
  #
18471
18826
  # resp = client.describe_job({
18472
18827
  # job_name: "JobName", # required
18473
- # job_category: "AgentRFT", # required, accepts AgentRFT, AgentRFTEvaluation
18828
+ # job_category: "AgentRFT", # required, accepts AgentRFT, SyntheticDataGeneration, DataQualityEvaluation, AgentRFTEvaluation, DataTransformation
18474
18829
  # })
18475
18830
  #
18476
18831
  # @example Response structure
@@ -18478,16 +18833,16 @@ module Aws::SageMaker
18478
18833
  # resp.job_name #=> String
18479
18834
  # resp.job_arn #=> String
18480
18835
  # resp.role_arn #=> String
18481
- # resp.job_category #=> String, one of "AgentRFT", "AgentRFTEvaluation"
18836
+ # resp.job_category #=> String, one of "AgentRFT", "SyntheticDataGeneration", "DataQualityEvaluation", "AgentRFTEvaluation", "DataTransformation"
18482
18837
  # resp.job_config_schema_version #=> String
18483
18838
  # resp.job_config_document #=> String
18484
18839
  # resp.creation_time #=> Time
18485
18840
  # resp.last_modified_time #=> Time
18486
18841
  # resp.end_time #=> Time
18487
18842
  # resp.job_status #=> String, one of "InProgress", "Completed", "Failed", "Stopping", "Stopped", "Deleting", "DeleteFailed"
18488
- # resp.secondary_status #=> String, one of "Starting", "Downloading", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Interrupted", "Failed", "Completed", "Restarting", "Pending", "Evaluating", "Deleting", "DeleteFailed"
18843
+ # resp.secondary_status #=> String, one of "Starting", "Downloading", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Interrupted", "Failed", "Completed", "Restarting", "Pending", "Generating", "PartialGeneration", "Evaluating", "Validating", "Preparing", "Executing", "Publishing", "Finalizing", "Transforming", "Deleting", "DeleteFailed"
18489
18844
  # resp.secondary_status_transitions #=> Array
18490
- # resp.secondary_status_transitions[0].status #=> String, one of "Starting", "Downloading", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Interrupted", "Failed", "Completed", "Restarting", "Pending", "Evaluating", "Deleting", "DeleteFailed"
18845
+ # resp.secondary_status_transitions[0].status #=> String, one of "Starting", "Downloading", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Interrupted", "Failed", "Completed", "Restarting", "Pending", "Generating", "PartialGeneration", "Evaluating", "Validating", "Preparing", "Executing", "Publishing", "Finalizing", "Transforming", "Deleting", "DeleteFailed"
18491
18846
  # resp.secondary_status_transitions[0].start_time #=> Time
18492
18847
  # resp.secondary_status_transitions[0].end_time #=> Time
18493
18848
  # resp.secondary_status_transitions[0].status_message #=> String
@@ -18533,13 +18888,13 @@ module Aws::SageMaker
18533
18888
  # @example Request syntax with placeholder values
18534
18889
  #
18535
18890
  # resp = client.describe_job_schema_version({
18536
- # job_category: "AgentRFT", # required, accepts AgentRFT, AgentRFTEvaluation
18891
+ # job_category: "AgentRFT", # required, accepts AgentRFT, SyntheticDataGeneration, DataQualityEvaluation, AgentRFTEvaluation, DataTransformation
18537
18892
  # job_config_schema_version: "JobSchemaVersion",
18538
18893
  # })
18539
18894
  #
18540
18895
  # @example Response structure
18541
18896
  #
18542
- # resp.job_category #=> String, one of "AgentRFT", "AgentRFTEvaluation"
18897
+ # resp.job_category #=> String, one of "AgentRFT", "SyntheticDataGeneration", "DataQualityEvaluation", "AgentRFTEvaluation", "DataTransformation"
18543
18898
  # resp.job_config_schema_version #=> String
18544
18899
  # resp.job_config_schema #=> String
18545
18900
  #
@@ -22022,6 +22377,40 @@ module Aws::SageMaker
22022
22377
  req.send_request(options)
22023
22378
  end
22024
22379
 
22380
+ # Returns the resource-based policy that is attached to an endpoint. If
22381
+ # the endpoint doesn't have a policy, you get a `ValidationException`.
22382
+ #
22383
+ # @option params [required, String] :endpoint_name
22384
+ # The name of the endpoint. The endpoint must be in your account and in
22385
+ # the current Amazon Web Services Region.
22386
+ #
22387
+ # @return [Types::GetEndpointPolicyResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
22388
+ #
22389
+ # * {Types::GetEndpointPolicyResponse#endpoint_name #endpoint_name} => String
22390
+ # * {Types::GetEndpointPolicyResponse#endpoint_arn #endpoint_arn} => String
22391
+ # * {Types::GetEndpointPolicyResponse#resource_policy #resource_policy} => String
22392
+ #
22393
+ # @example Request syntax with placeholder values
22394
+ #
22395
+ # resp = client.get_endpoint_policy({
22396
+ # endpoint_name: "EndpointName", # required
22397
+ # })
22398
+ #
22399
+ # @example Response structure
22400
+ #
22401
+ # resp.endpoint_name #=> String
22402
+ # resp.endpoint_arn #=> String
22403
+ # resp.resource_policy #=> String
22404
+ #
22405
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/GetEndpointPolicy AWS API Documentation
22406
+ #
22407
+ # @overload get_endpoint_policy(params = {})
22408
+ # @param [Hash] params ({})
22409
+ def get_endpoint_policy(params = {}, options = {})
22410
+ req = build_request(:get_endpoint_policy, params)
22411
+ req.send_request(options)
22412
+ end
22413
+
22025
22414
  # The resource policy for the lineage group.
22026
22415
  #
22027
22416
  # @option params [required, String] :lineage_group_name
@@ -23275,6 +23664,12 @@ module Aws::SageMaker
23275
23664
  # The type of resource for which to filter events. Valid values are
23276
23665
  # `Cluster`, `InstanceGroup`, or `Instance`.
23277
23666
  #
23667
+ # @option params [String] :operation_id
23668
+ # The unique identifier (UUID) of a cluster operation for which to
23669
+ # filter events. If specified, only events associated with this
23670
+ # operation are returned. This ID can be obtained from the
23671
+ # `ListClusterOperations` operation.
23672
+ #
23278
23673
  # @option params [Integer] :max_results
23279
23674
  # The maximum number of events to return in the response. Valid range is
23280
23675
  # 1 to 100.
@@ -23301,6 +23696,7 @@ module Aws::SageMaker
23301
23696
  # sort_by: "EventTime", # accepts EventTime
23302
23697
  # sort_order: "Ascending", # accepts Ascending, Descending
23303
23698
  # resource_type: "Cluster", # accepts Cluster, InstanceGroup, Instance
23699
+ # operation_id: "ClusterOperationId",
23304
23700
  # max_results: 1,
23305
23701
  # next_token: "NextToken",
23306
23702
  # })
@@ -23318,6 +23714,7 @@ module Aws::SageMaker
23318
23714
  # resp.events[0].event_time #=> Time
23319
23715
  # resp.events[0].description #=> String
23320
23716
  # resp.events[0].event_level #=> String, one of "Info", "Warn", "Error"
23717
+ # resp.events[0].operation_id #=> String
23321
23718
  #
23322
23719
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ListClusterEvents AWS API Documentation
23323
23720
  #
@@ -23328,6 +23725,59 @@ module Aws::SageMaker
23328
23725
  req.send_request(options)
23329
23726
  end
23330
23727
 
23728
+ # Lists the versions of an extension that you can install on a SageMaker
23729
+ # HyperPod cluster, from the highest version to the lowest. Use a
23730
+ # version from this list as the `ExtensionVersion` in
23731
+ # `CreateClusterExtension` or `UpdateClusterExtension`.
23732
+ #
23733
+ # The list contains only versions that you can install. Versions that
23734
+ # SageMaker HyperPod no longer offers for new installations don't
23735
+ # appear. If SageMaker HyperPod doesn't offer an extension with the
23736
+ # name that you specify, the response contains an empty list.
23737
+ #
23738
+ # @option params [required, String] :extension_name
23739
+ # The name of the extension whose versions you want to list, for example
23740
+ # `sagemaker-observability`.
23741
+ #
23742
+ # @option params [Integer] :max_results
23743
+ # The maximum number of versions to return in the response. If you
23744
+ # don't specify a value, SageMaker HyperPod returns up to 100 versions.
23745
+ #
23746
+ # @option params [String] :next_token
23747
+ # If the result of the previous `ListClusterExtensionVersions` request
23748
+ # was truncated, the response includes a `NextToken`. To retrieve the
23749
+ # next set of versions, use the token in the next request.
23750
+ #
23751
+ # @return [Types::ListClusterExtensionVersionsResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
23752
+ #
23753
+ # * {Types::ListClusterExtensionVersionsResponse#extension_versions #extension_versions} => Array&lt;Types::ExtensionVersionSummary&gt;
23754
+ # * {Types::ListClusterExtensionVersionsResponse#next_token #next_token} => String
23755
+ #
23756
+ # The returned {Seahorse::Client::Response response} is a pageable response and is Enumerable. For details on usage see {Aws::PageableResponse PageableResponse}.
23757
+ #
23758
+ # @example Request syntax with placeholder values
23759
+ #
23760
+ # resp = client.list_cluster_extension_versions({
23761
+ # extension_name: "ExtensionName", # required
23762
+ # max_results: 1,
23763
+ # next_token: "NextToken",
23764
+ # })
23765
+ #
23766
+ # @example Response structure
23767
+ #
23768
+ # resp.extension_versions #=> Array
23769
+ # resp.extension_versions[0].extension_version #=> String
23770
+ # resp.next_token #=> String
23771
+ #
23772
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ListClusterExtensionVersions AWS API Documentation
23773
+ #
23774
+ # @overload list_cluster_extension_versions(params = {})
23775
+ # @param [Hash] params ({})
23776
+ def list_cluster_extension_versions(params = {}, options = {})
23777
+ req = build_request(:list_cluster_extension_versions, params)
23778
+ req.send_request(options)
23779
+ end
23780
+
23331
23781
  # Lists the extensions installed on a SageMaker HyperPod cluster.
23332
23782
  #
23333
23783
  # @option params [required, String] :cluster_arn
@@ -23510,6 +23960,87 @@ module Aws::SageMaker
23510
23960
  req.send_request(options)
23511
23961
  end
23512
23962
 
23963
+ # Retrieves a list of operation summaries for a specified HyperPod
23964
+ # cluster, such as scaling operations and deep health checks. The
23965
+ # operation supports filtering, sorting, and pagination of results.
23966
+ #
23967
+ # @option params [required, String] :cluster_name
23968
+ # The name or Amazon Resource Name (ARN) of the HyperPod cluster for
23969
+ # which to list operations.
23970
+ #
23971
+ # @option params [String] :status
23972
+ # The operation status to filter by. If specified, only operations with
23973
+ # this status are returned. Valid values are `InProgress`, `Superseded`,
23974
+ # or `Done`.
23975
+ #
23976
+ # @option params [String] :type
23977
+ # The operation type to filter by. If specified, only operations of this
23978
+ # type are returned. Valid values are `Scaling` or `DeepHealthCheck`.
23979
+ #
23980
+ # @option params [Time,DateTime,Date,Integer,String] :creation_time_after
23981
+ # The start of the time range for filtering operations. Only operations
23982
+ # created after this time are included in the results.
23983
+ #
23984
+ # @option params [Time,DateTime,Date,Integer,String] :creation_time_before
23985
+ # The end of the time range for filtering operations. Only operations
23986
+ # created before this time are included in the results.
23987
+ #
23988
+ # @option params [String] :sort_by
23989
+ # The field to use for sorting the operation list. Currently, the only
23990
+ # supported value is `CreationTime`.
23991
+ #
23992
+ # @option params [String] :sort_order
23993
+ # The order in which to sort the results. Valid values are `Ascending`
23994
+ # or `Descending` (the default is `Descending`).
23995
+ #
23996
+ # @option params [Integer] :max_results
23997
+ # The maximum number of operations to return in the response. Valid
23998
+ # range is 1 to 100.
23999
+ #
24000
+ # @option params [String] :next_token
24001
+ # A token to retrieve the next set of results. This token is obtained
24002
+ # from the output of a previous `ListClusterOperations` call.
24003
+ #
24004
+ # @return [Types::ListClusterOperationsResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
24005
+ #
24006
+ # * {Types::ListClusterOperationsResponse#next_token #next_token} => String
24007
+ # * {Types::ListClusterOperationsResponse#cluster_operation_summaries #cluster_operation_summaries} => Array&lt;Types::ClusterOperationSummary&gt;
24008
+ #
24009
+ # The returned {Seahorse::Client::Response response} is a pageable response and is Enumerable. For details on usage see {Aws::PageableResponse PageableResponse}.
24010
+ #
24011
+ # @example Request syntax with placeholder values
24012
+ #
24013
+ # resp = client.list_cluster_operations({
24014
+ # cluster_name: "ClusterNameOrArn", # required
24015
+ # status: "InProgress", # accepts InProgress, Superseded, Done
24016
+ # type: "Scaling", # accepts Scaling, DeepHealthCheck
24017
+ # creation_time_after: Time.now,
24018
+ # creation_time_before: Time.now,
24019
+ # sort_by: "CreationTime", # accepts CreationTime
24020
+ # sort_order: "Ascending", # accepts Ascending, Descending
24021
+ # max_results: 1,
24022
+ # next_token: "NextToken",
24023
+ # })
24024
+ #
24025
+ # @example Response structure
24026
+ #
24027
+ # resp.next_token #=> String
24028
+ # resp.cluster_operation_summaries #=> Array
24029
+ # resp.cluster_operation_summaries[0].operation_id #=> String
24030
+ # resp.cluster_operation_summaries[0].cluster_arn #=> String
24031
+ # resp.cluster_operation_summaries[0].type #=> String, one of "Scaling", "DeepHealthCheck"
24032
+ # resp.cluster_operation_summaries[0].status #=> String, one of "InProgress", "Superseded", "Done"
24033
+ # resp.cluster_operation_summaries[0].creation_time #=> Time
24034
+ #
24035
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ListClusterOperations AWS API Documentation
24036
+ #
24037
+ # @overload list_cluster_operations(params = {})
24038
+ # @param [Hash] params ({})
24039
+ def list_cluster_operations(params = {}, options = {})
24040
+ req = build_request(:list_cluster_operations, params)
24041
+ req.send_request(options)
24042
+ end
24043
+
23513
24044
  # List the cluster policy configurations.
23514
24045
  #
23515
24046
  # @option params [Time,DateTime,Date,Integer,String] :created_after
@@ -25765,6 +26296,49 @@ module Aws::SageMaker
25765
26296
  req.send_request(options)
25766
26297
  end
25767
26298
 
26299
+ # Lists the inference routers in your account in the current Amazon Web
26300
+ # Services Region. A router that you just created might not appear in
26301
+ # the list right away.
26302
+ #
26303
+ # @option params [Integer] :max_results
26304
+ # The maximum number of inference routers to return in the response.
26305
+ #
26306
+ # @option params [String] :next_token
26307
+ # If the previous response was truncated, you receive this token. To get
26308
+ # the next set of inference routers, use it in your next request.
26309
+ #
26310
+ # @return [Types::ListInferenceRoutersOutput] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
26311
+ #
26312
+ # * {Types::ListInferenceRoutersOutput#inference_routers #inference_routers} => Array&lt;Types::InferenceRouterSummary&gt;
26313
+ # * {Types::ListInferenceRoutersOutput#next_token #next_token} => String
26314
+ #
26315
+ # The returned {Seahorse::Client::Response response} is a pageable response and is Enumerable. For details on usage see {Aws::PageableResponse PageableResponse}.
26316
+ #
26317
+ # @example Request syntax with placeholder values
26318
+ #
26319
+ # resp = client.list_inference_routers({
26320
+ # max_results: 1,
26321
+ # next_token: "NextToken",
26322
+ # })
26323
+ #
26324
+ # @example Response structure
26325
+ #
26326
+ # resp.inference_routers #=> Array
26327
+ # resp.inference_routers[0].inference_router_name #=> String
26328
+ # resp.inference_routers[0].inference_router_arn #=> String
26329
+ # resp.inference_routers[0].creation_time #=> Time
26330
+ # resp.inference_routers[0].last_modified_time #=> Time
26331
+ # resp.next_token #=> String
26332
+ #
26333
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ListInferenceRouters AWS API Documentation
26334
+ #
26335
+ # @overload list_inference_routers(params = {})
26336
+ # @param [Hash] params ({})
26337
+ def list_inference_routers(params = {}, options = {})
26338
+ req = build_request(:list_inference_routers, params)
26339
+ req.send_request(options)
26340
+ end
26341
+
25768
26342
  # Lists available configuration schema versions for a specified job
25769
26343
  # category. Use the schema versions with `DescribeJobSchemaVersion` to
25770
26344
  # retrieve the full schema document.
@@ -25796,7 +26370,7 @@ module Aws::SageMaker
25796
26370
  # @example Request syntax with placeholder values
25797
26371
  #
25798
26372
  # resp = client.list_job_schema_versions({
25799
- # job_category: "AgentRFT", # required, accepts AgentRFT, AgentRFTEvaluation
26373
+ # job_category: "AgentRFT", # required, accepts AgentRFT, SyntheticDataGeneration, DataQualityEvaluation, AgentRFTEvaluation, DataTransformation
25800
26374
  # next_token: "NextToken",
25801
26375
  # max_results: 1,
25802
26376
  # })
@@ -25874,7 +26448,7 @@ module Aws::SageMaker
25874
26448
  # @example Request syntax with placeholder values
25875
26449
  #
25876
26450
  # resp = client.list_jobs({
25877
- # job_category: "AgentRFT", # required, accepts AgentRFT, AgentRFTEvaluation
26451
+ # job_category: "AgentRFT", # required, accepts AgentRFT, SyntheticDataGeneration, DataQualityEvaluation, AgentRFTEvaluation, DataTransformation
25878
26452
  # next_token: "NextToken",
25879
26453
  # max_results: 1,
25880
26454
  # creation_time_after: Time.now,
@@ -25893,9 +26467,9 @@ module Aws::SageMaker
25893
26467
  # resp.job_summaries #=> Array
25894
26468
  # resp.job_summaries[0].job_arn #=> String
25895
26469
  # resp.job_summaries[0].job_name #=> String
25896
- # resp.job_summaries[0].job_category #=> String, one of "AgentRFT", "AgentRFTEvaluation"
26470
+ # resp.job_summaries[0].job_category #=> String, one of "AgentRFT", "SyntheticDataGeneration", "DataQualityEvaluation", "AgentRFTEvaluation", "DataTransformation"
25897
26471
  # resp.job_summaries[0].job_status #=> String, one of "InProgress", "Completed", "Failed", "Stopping", "Stopped", "Deleting", "DeleteFailed"
25898
- # resp.job_summaries[0].job_secondary_status #=> String, one of "Starting", "Downloading", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Interrupted", "Failed", "Completed", "Restarting", "Pending", "Evaluating", "Deleting", "DeleteFailed"
26472
+ # resp.job_summaries[0].job_secondary_status #=> String, one of "Starting", "Downloading", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Interrupted", "Failed", "Completed", "Restarting", "Pending", "Generating", "PartialGeneration", "Evaluating", "Validating", "Preparing", "Executing", "Publishing", "Finalizing", "Transforming", "Deleting", "DeleteFailed"
25899
26473
  # resp.job_summaries[0].creation_time #=> Time
25900
26474
  # resp.job_summaries[0].last_modified_time #=> Time
25901
26475
  # resp.job_summaries[0].end_time #=> Time
@@ -29543,6 +30117,62 @@ module Aws::SageMaker
29543
30117
  req.send_request(options)
29544
30118
  end
29545
30119
 
30120
+ # Attaches a resource-based policy to an endpoint. With an endpoint
30121
+ # policy, you can allow other Amazon Web Services accounts to send
30122
+ # requests to your endpoint, or to inference components on it, through
30123
+ # an inference router. The policy that you specify replaces the
30124
+ # endpoint's existing policy, if it has one.
30125
+ #
30126
+ # You can't attach a policy to an endpoint whose status is `Deleting`
30127
+ # or `Failed`. SageMaker AI rejects a policy that doesn't follow the
30128
+ # endpoint policy rules. For information about the policy rules, see
30129
+ # [Use destinations in other accounts][1].
30130
+ #
30131
+ # Policy changes typically take effect within 30 seconds, and can take
30132
+ # up to 5 minutes. If you delete the endpoint, its policy is also
30133
+ # deleted, and a new endpoint with the same name has no policy.
30134
+ #
30135
+ #
30136
+ #
30137
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/inference-router-cross-account.html
30138
+ #
30139
+ # @option params [required, String] :endpoint_name
30140
+ # The name of the endpoint. The endpoint must be in your account and in
30141
+ # the current Amazon Web Services Region.
30142
+ #
30143
+ # @option params [required, String] :resource_policy
30144
+ # The endpoint policy, as a JSON policy document. The policy can contain
30145
+ # only ASCII characters. For information about the policy rules, see
30146
+ # [Use destinations in other accounts][1].
30147
+ #
30148
+ #
30149
+ #
30150
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/inference-router-cross-account.html
30151
+ #
30152
+ # @return [Types::PutEndpointPolicyResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
30153
+ #
30154
+ # * {Types::PutEndpointPolicyResponse#endpoint_arn #endpoint_arn} => String
30155
+ #
30156
+ # @example Request syntax with placeholder values
30157
+ #
30158
+ # resp = client.put_endpoint_policy({
30159
+ # endpoint_name: "EndpointName", # required
30160
+ # resource_policy: "EndpointPolicyDocument", # required
30161
+ # })
30162
+ #
30163
+ # @example Response structure
30164
+ #
30165
+ # resp.endpoint_arn #=> String
30166
+ #
30167
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/PutEndpointPolicy AWS API Documentation
30168
+ #
30169
+ # @overload put_endpoint_policy(params = {})
30170
+ # @param [Hash] params ({})
30171
+ def put_endpoint_policy(params = {}, options = {})
30172
+ req = build_request(:put_endpoint_policy, params)
30173
+ req.send_request(options)
30174
+ end
30175
+
29546
30176
  # Adds a resouce policy to control access to a model group. For
29547
30177
  # information about resoure policies, see [Identity-based policies and
29548
30178
  # resource-based policies][1] in the *Amazon Web Services Identity and
@@ -30803,7 +31433,7 @@ module Aws::SageMaker
30803
31433
  #
30804
31434
  # resp = client.stop_job({
30805
31435
  # job_name: "JobName", # required
30806
- # job_category: "AgentRFT", # required, accepts AgentRFT, AgentRFTEvaluation
31436
+ # job_category: "AgentRFT", # required, accepts AgentRFT, SyntheticDataGeneration, DataQualityEvaluation, AgentRFTEvaluation, DataTransformation
30807
31437
  # })
30808
31438
  #
30809
31439
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/StopJob AWS API Documentation
@@ -31405,7 +32035,7 @@ module Aws::SageMaker
31405
32035
  # },
31406
32036
  # image_id: "ImageId",
31407
32037
  # auto_patch_config: {
31408
- # patching_strategy: "WhenIdle", # required, accepts WhenIdle, WhenAllIdle
32038
+ # patching_strategy: "WhenIdle", # required, accepts WhenIdle, WhenAllIdle, DrainFirst
31409
32039
  # patch_schedule: {
31410
32040
  # next_patch_date: Time.now,
31411
32041
  # },
@@ -31427,6 +32057,11 @@ module Aws::SageMaker
31427
32057
  # },
31428
32058
  # ],
31429
32059
  # },
32060
+ # idle_wait_window_in_seconds: 1,
32061
+ # drain_policy: {
32062
+ # drain_timeout_in_seconds: 1,
32063
+ # on_drain_timeout: "Force", # accepts Force, Skip
32064
+ # },
31430
32065
  # },
31431
32066
  # image_release_version: "ImageReleaseVersion",
31432
32067
  # kubernetes_config: {
@@ -31580,6 +32215,11 @@ module Aws::SageMaker
31580
32215
  # an earlier version. To use an earlier version, delete the extension
31581
32216
  # and create it again with that version.
31582
32217
  #
32218
+ # To retry the operation on nodes where it failed, fix the cause that
32219
+ # `DescribeClusterExtension` reports in `Health`, and then send an
32220
+ # `UpdateClusterExtension` request that doesn't change the version or
32221
+ # the configuration.
32222
+ #
31583
32223
  # You can't update an extension while it is being deleted.
31584
32224
  #
31585
32225
  # @option params [required, String] :cluster_arn
@@ -33477,6 +34117,80 @@ module Aws::SageMaker
33477
34117
  req.send_request(options)
33478
34118
  end
33479
34119
 
34120
+ # Updates the routing targets of an inference router. The routing
34121
+ # targets that you specify replace all of the router's existing routing
34122
+ # targets. To change one routing target, call
34123
+ # [DescribeInferenceRouter][1], edit the routing targets, and send all
34124
+ # of them.
34125
+ #
34126
+ # The update is applied in full or not at all, and each request uses
34127
+ # either the previous configuration or the new one. Updates typically
34128
+ # take effect within about 2 minutes.
34129
+ #
34130
+ # If the router is being created, updated, or deleted, or another
34131
+ # request changes it at the same time, you get a `ConflictException`.
34132
+ # Describe the router, and then send your update again.
34133
+ #
34134
+ #
34135
+ #
34136
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeInferenceRouter.html
34137
+ #
34138
+ # @option params [required, String] :inference_router_name
34139
+ # The name of the inference router to update.
34140
+ #
34141
+ # @option params [required, Array<Types::RoutingTarget>] :routing_targets
34142
+ # The complete list of routing targets for the inference router. To
34143
+ # remove the last routing target, delete the router.
34144
+ #
34145
+ # @return [Types::UpdateInferenceRouterOutput] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
34146
+ #
34147
+ # * {Types::UpdateInferenceRouterOutput#inference_router_arn #inference_router_arn} => String
34148
+ #
34149
+ # @example Request syntax with placeholder values
34150
+ #
34151
+ # resp = client.update_inference_router({
34152
+ # inference_router_name: "InferenceRouterName", # required
34153
+ # routing_targets: [ # required
34154
+ # {
34155
+ # name: "RoutingTargetName", # required
34156
+ # destinations: [ # required
34157
+ # {
34158
+ # resource_arn: "DestinationResourceArn", # required
34159
+ # routing_status: "Routable", # accepts Routable, Unroutable, Unknown
34160
+ # },
34161
+ # ],
34162
+ # async_inference_config: {
34163
+ # output_config: { # required
34164
+ # s3_output_path: "InferenceRouterS3Uri", # required
34165
+ # s3_failure_path: "InferenceRouterS3Uri", # required
34166
+ # kms_key_id: "KmsKeyId",
34167
+ # notification_config: {
34168
+ # success_topic: "InferenceRouterSnsTopicArn",
34169
+ # error_topic: "InferenceRouterSnsTopicArn",
34170
+ # },
34171
+ # },
34172
+ # client_config: {
34173
+ # max_time_in_backlog_seconds: 1,
34174
+ # },
34175
+ # },
34176
+ # routing_preference: "LocalRegionFirst", # accepts LocalRegionFirst, InstancePoolOrder
34177
+ # },
34178
+ # ],
34179
+ # })
34180
+ #
34181
+ # @example Response structure
34182
+ #
34183
+ # resp.inference_router_arn #=> String
34184
+ #
34185
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/UpdateInferenceRouter AWS API Documentation
34186
+ #
34187
+ # @overload update_inference_router(params = {})
34188
+ # @param [Hash] params ({})
34189
+ def update_inference_router(params = {}, options = {})
34190
+ req = build_request(:update_inference_router, params)
34191
+ req.send_request(options)
34192
+ end
34193
+
33480
34194
  # Updates an MLflow App.
33481
34195
  #
33482
34196
  # @option params [required, String] :arn
@@ -35579,7 +36293,7 @@ module Aws::SageMaker
35579
36293
  tracer: tracer
35580
36294
  )
35581
36295
  context[:gem_name] = 'aws-sdk-sagemaker'
35582
- context[:gem_version] = '1.414.0'
36296
+ context[:gem_version] = '1.416.0'
35583
36297
  Seahorse::Client::Request.new(handlers, context)
35584
36298
  end
35585
36299