aws-sdk-sagemaker 1.414.0 → 1.415.0

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.
checksums.yaml CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA256:
3
- metadata.gz: 4dfa95f7a7bfa274e1bdcafdbb47c31be8c40a5caade1d18cfd8aa65c73bbe2a
4
- data.tar.gz: 8a5fcd38cb7bee7d0eb5eb4c5f970c8432b91d732d55dd7d0018a0599f98d8fd
3
+ metadata.gz: 48a9dbca80856e37ced930df604856b9f7abb6e32b62cf7b08c949885bc6ea16
4
+ data.tar.gz: fd20d262709b54cf44b3bdfcf08d2bbb67e01f47e6d1cfd1526c135b347789b1
5
5
  SHA512:
6
- metadata.gz: 19f46d963f4bb3c05ccf90089892d41f1617532f8e33666cb232429577b06ff93d98f08e90056dba23131a547584aa4ec84d47469d3d6e3bd685748048bce674
7
- data.tar.gz: 12603081f3e1386622fe5d5248864966dfa3acf49b401f2774785de1bca6cb3d2a0c1649dc5a03ec596078abf596e4992c165e3822f395d58e007069c6c288b7
6
+ metadata.gz: daa018b3671fcdd6f646620ce3fc977ddd21ef1ba87c3588a0bb0e5d3c9fa5fef66384bd9d9355286ea8478f16d5a201c07442db0be1585bfd0873cc3c7c3862
7
+ data.tar.gz: fd67f58343b12b89f3962fc936704c01c92519e4f64e1e96c6f184729bf441d26edadacf5128f653434f30afe4d150a13e7a801c4b600b8fb34f6b4a27e527b3
data/CHANGELOG.md CHANGED
@@ -1,6 +1,11 @@
1
1
  Unreleased Changes
2
2
  ------------------
3
3
 
4
+ 1.415.0 (2026-11-13)
5
+ ------------------
6
+
7
+ * Feature - New feature for Sagemaker Hyperpod Slurm clusters to install Extensions and view cluster operations.
8
+
4
9
  1.414.0 (2026-11-12)
5
10
  ------------------
6
11
 
data/VERSION CHANGED
@@ -1 +1 @@
1
- 1.414.0
1
+ 1.415.0
@@ -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: {
@@ -15482,7 +15487,7 @@ module Aws::SageMaker
15482
15487
  # resp.instance_groups[0].scheduled_update_config.deployment_config.wait_interval_in_seconds #=> Integer
15483
15488
  # resp.instance_groups[0].scheduled_update_config.deployment_config.auto_rollback_configuration #=> Array
15484
15489
  # 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"
15490
+ # resp.instance_groups[0].auto_patch_config.patching_strategy #=> String, one of "WhenIdle", "WhenAllIdle", "DrainFirst"
15486
15491
  # resp.instance_groups[0].auto_patch_config.current_patch_schedule.next_patch_date #=> Time
15487
15492
  # resp.instance_groups[0].auto_patch_config.desired_patch_schedule.next_patch_date #=> Time
15488
15493
  # 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 +15497,9 @@ module Aws::SageMaker
15492
15497
  # resp.instance_groups[0].auto_patch_config.deployment_config.wait_interval_in_seconds #=> Integer
15493
15498
  # resp.instance_groups[0].auto_patch_config.deployment_config.auto_rollback_configuration #=> Array
15494
15499
  # resp.instance_groups[0].auto_patch_config.deployment_config.auto_rollback_configuration[0].alarm_name #=> String
15500
+ # resp.instance_groups[0].auto_patch_config.idle_wait_window_in_seconds #=> Integer
15501
+ # resp.instance_groups[0].auto_patch_config.drain_policy.drain_timeout_in_seconds #=> Integer
15502
+ # resp.instance_groups[0].auto_patch_config.drain_policy.on_drain_timeout #=> String, one of "Force", "Skip"
15495
15503
  # resp.instance_groups[0].current_image_id #=> String
15496
15504
  # resp.instance_groups[0].desired_image_id #=> String
15497
15505
  # resp.instance_groups[0].current_image_release_version #=> String
@@ -15663,6 +15671,7 @@ module Aws::SageMaker
15663
15671
  # resp.event_details.event_details.event_metadata.slurm_health.reason #=> String, one of "DaemonDown", "DaemonDisabled", "DbUnreachable"
15664
15672
  # resp.event_details.description #=> String
15665
15673
  # resp.event_details.event_level #=> String, one of "Info", "Warn", "Error"
15674
+ # resp.event_details.operation_id #=> String
15666
15675
  #
15667
15676
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeClusterEvent AWS API Documentation
15668
15677
  #
@@ -15676,7 +15685,9 @@ module Aws::SageMaker
15676
15685
  # Retrieves information about an extension installed on a SageMaker
15677
15686
  # HyperPod cluster. The information includes the extension's version,
15678
15687
  # configuration, and status, and a summary of how many cluster nodes are
15679
- # running the current version and configuration of the extension.
15688
+ # running the current version and configuration of the extension. If the
15689
+ # extension has problems on some nodes, the response also lists the
15690
+ # problems and how many nodes each one affects.
15680
15691
  #
15681
15692
  # @option params [required, String] :cluster_arn
15682
15693
  # The Amazon Resource Name (ARN) of the SageMaker HyperPod cluster on
@@ -15692,6 +15703,7 @@ module Aws::SageMaker
15692
15703
  #
15693
15704
  # * {Types::DescribeClusterExtensionResponse#extension_details #extension_details} => Types::ExtensionDetails
15694
15705
  # * {Types::DescribeClusterExtensionResponse#node_summary #node_summary} => Types::ExtensionNodeSummary
15706
+ # * {Types::DescribeClusterExtensionResponse#health #health} => Types::ExtensionHealth
15695
15707
  #
15696
15708
  # @example Request syntax with placeholder values
15697
15709
  #
@@ -15712,6 +15724,10 @@ module Aws::SageMaker
15712
15724
  # resp.extension_details.last_modified_time #=> Time
15713
15725
  # resp.node_summary.total_count #=> Integer
15714
15726
  # resp.node_summary.active_count #=> Integer
15727
+ # resp.health.issues #=> Array
15728
+ # resp.health.issues[0].code #=> String, one of "InvalidConfiguration", "ConfigurationConflict", "AccessDenied", "ResourceNotFound", "DependencyUnreachable", "ExtensionUnhealthy", "InternalFailure", "Other"
15729
+ # resp.health.issues[0].message #=> String
15730
+ # resp.health.issues[0].node_count #=> Integer
15715
15731
  #
15716
15732
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeClusterExtension AWS API Documentation
15717
15733
  #
@@ -15813,6 +15829,68 @@ module Aws::SageMaker
15813
15829
  req.send_request(options)
15814
15830
  end
15815
15831
 
15832
+ # Retrieves detailed information about a specific operation for a given
15833
+ # HyperPod cluster, such as a scaling operation or a deep health check.
15834
+ # Use the `OperationId` returned by the `ListClusterOperations`
15835
+ # operation to identify the operation to describe.
15836
+ #
15837
+ # @option params [required, String] :cluster_name
15838
+ # The name or Amazon Resource Name (ARN) of the HyperPod cluster
15839
+ # associated with the operation.
15840
+ #
15841
+ # @option params [required, String] :operation_id
15842
+ # The unique identifier (UUID) of the operation to describe. This ID can
15843
+ # be obtained from the `ListClusterOperations` operation.
15844
+ #
15845
+ # @return [Types::DescribeClusterOperationResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
15846
+ #
15847
+ # * {Types::DescribeClusterOperationResponse#operation_id #operation_id} => String
15848
+ # * {Types::DescribeClusterOperationResponse#cluster_arn #cluster_arn} => String
15849
+ # * {Types::DescribeClusterOperationResponse#type #type} => String
15850
+ # * {Types::DescribeClusterOperationResponse#status #status} => String
15851
+ # * {Types::DescribeClusterOperationResponse#creation_time #creation_time} => Time
15852
+ # * {Types::DescribeClusterOperationResponse#scaling_details #scaling_details} => Array<Types::ClusterOperationScalingDetails>
15853
+ # * {Types::DescribeClusterOperationResponse#deep_health_check_details #deep_health_check_details} => Array<Types::InstanceGroupHealthCheckConfiguration>
15854
+ #
15855
+ # @example Request syntax with placeholder values
15856
+ #
15857
+ # resp = client.describe_cluster_operation({
15858
+ # cluster_name: "ClusterNameOrArn", # required
15859
+ # operation_id: "ClusterOperationId", # required
15860
+ # })
15861
+ #
15862
+ # @example Response structure
15863
+ #
15864
+ # resp.operation_id #=> String
15865
+ # resp.cluster_arn #=> String
15866
+ # resp.type #=> String, one of "Scaling", "DeepHealthCheck"
15867
+ # resp.status #=> String, one of "InProgress", "Superseded", "Done"
15868
+ # resp.creation_time #=> Time
15869
+ # resp.scaling_details #=> Array
15870
+ # resp.scaling_details[0].instance_group_name #=> String
15871
+ # resp.scaling_details[0].status #=> String, one of "InProgress", "Superseded", "Done"
15872
+ # resp.scaling_details[0].original.current_count #=> Integer
15873
+ # resp.scaling_details[0].original.target_count #=> Integer
15874
+ # resp.scaling_details[0].original.min_count #=> Integer
15875
+ # resp.scaling_details[0].desired.current_count #=> Integer
15876
+ # resp.scaling_details[0].desired.target_count #=> Integer
15877
+ # resp.scaling_details[0].desired.min_count #=> Integer
15878
+ # resp.deep_health_check_details #=> Array
15879
+ # resp.deep_health_check_details[0].instance_group_name #=> String
15880
+ # resp.deep_health_check_details[0].instance_ids #=> Array
15881
+ # resp.deep_health_check_details[0].instance_ids[0] #=> String
15882
+ # resp.deep_health_check_details[0].deep_health_checks #=> Array
15883
+ # resp.deep_health_check_details[0].deep_health_checks[0] #=> String, one of "InstanceStress", "InstanceConnectivity", "MultiNodeTrainingValidation"
15884
+ #
15885
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeClusterOperation AWS API Documentation
15886
+ #
15887
+ # @overload describe_cluster_operation(params = {})
15888
+ # @param [Hash] params ({})
15889
+ def describe_cluster_operation(params = {}, options = {})
15890
+ req = build_request(:describe_cluster_operation, params)
15891
+ req.send_request(options)
15892
+ end
15893
+
15816
15894
  # Description of the cluster policy. This policy is used for task
15817
15895
  # prioritization and fair-share allocation. This helps prioritize
15818
15896
  # critical workloads and distributes idle compute across entities.
@@ -23275,6 +23353,12 @@ module Aws::SageMaker
23275
23353
  # The type of resource for which to filter events. Valid values are
23276
23354
  # `Cluster`, `InstanceGroup`, or `Instance`.
23277
23355
  #
23356
+ # @option params [String] :operation_id
23357
+ # The unique identifier (UUID) of a cluster operation for which to
23358
+ # filter events. If specified, only events associated with this
23359
+ # operation are returned. This ID can be obtained from the
23360
+ # `ListClusterOperations` operation.
23361
+ #
23278
23362
  # @option params [Integer] :max_results
23279
23363
  # The maximum number of events to return in the response. Valid range is
23280
23364
  # 1 to 100.
@@ -23301,6 +23385,7 @@ module Aws::SageMaker
23301
23385
  # sort_by: "EventTime", # accepts EventTime
23302
23386
  # sort_order: "Ascending", # accepts Ascending, Descending
23303
23387
  # resource_type: "Cluster", # accepts Cluster, InstanceGroup, Instance
23388
+ # operation_id: "ClusterOperationId",
23304
23389
  # max_results: 1,
23305
23390
  # next_token: "NextToken",
23306
23391
  # })
@@ -23318,6 +23403,7 @@ module Aws::SageMaker
23318
23403
  # resp.events[0].event_time #=> Time
23319
23404
  # resp.events[0].description #=> String
23320
23405
  # resp.events[0].event_level #=> String, one of "Info", "Warn", "Error"
23406
+ # resp.events[0].operation_id #=> String
23321
23407
  #
23322
23408
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ListClusterEvents AWS API Documentation
23323
23409
  #
@@ -23328,6 +23414,59 @@ module Aws::SageMaker
23328
23414
  req.send_request(options)
23329
23415
  end
23330
23416
 
23417
+ # Lists the versions of an extension that you can install on a SageMaker
23418
+ # HyperPod cluster, from the highest version to the lowest. Use a
23419
+ # version from this list as the `ExtensionVersion` in
23420
+ # `CreateClusterExtension` or `UpdateClusterExtension`.
23421
+ #
23422
+ # The list contains only versions that you can install. Versions that
23423
+ # SageMaker HyperPod no longer offers for new installations don't
23424
+ # appear. If SageMaker HyperPod doesn't offer an extension with the
23425
+ # name that you specify, the response contains an empty list.
23426
+ #
23427
+ # @option params [required, String] :extension_name
23428
+ # The name of the extension whose versions you want to list, for example
23429
+ # `sagemaker-observability`.
23430
+ #
23431
+ # @option params [Integer] :max_results
23432
+ # The maximum number of versions to return in the response. If you
23433
+ # don't specify a value, SageMaker HyperPod returns up to 100 versions.
23434
+ #
23435
+ # @option params [String] :next_token
23436
+ # If the result of the previous `ListClusterExtensionVersions` request
23437
+ # was truncated, the response includes a `NextToken`. To retrieve the
23438
+ # next set of versions, use the token in the next request.
23439
+ #
23440
+ # @return [Types::ListClusterExtensionVersionsResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
23441
+ #
23442
+ # * {Types::ListClusterExtensionVersionsResponse#extension_versions #extension_versions} => Array<Types::ExtensionVersionSummary>
23443
+ # * {Types::ListClusterExtensionVersionsResponse#next_token #next_token} => String
23444
+ #
23445
+ # The returned {Seahorse::Client::Response response} is a pageable response and is Enumerable. For details on usage see {Aws::PageableResponse PageableResponse}.
23446
+ #
23447
+ # @example Request syntax with placeholder values
23448
+ #
23449
+ # resp = client.list_cluster_extension_versions({
23450
+ # extension_name: "ExtensionName", # required
23451
+ # max_results: 1,
23452
+ # next_token: "NextToken",
23453
+ # })
23454
+ #
23455
+ # @example Response structure
23456
+ #
23457
+ # resp.extension_versions #=> Array
23458
+ # resp.extension_versions[0].extension_version #=> String
23459
+ # resp.next_token #=> String
23460
+ #
23461
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ListClusterExtensionVersions AWS API Documentation
23462
+ #
23463
+ # @overload list_cluster_extension_versions(params = {})
23464
+ # @param [Hash] params ({})
23465
+ def list_cluster_extension_versions(params = {}, options = {})
23466
+ req = build_request(:list_cluster_extension_versions, params)
23467
+ req.send_request(options)
23468
+ end
23469
+
23331
23470
  # Lists the extensions installed on a SageMaker HyperPod cluster.
23332
23471
  #
23333
23472
  # @option params [required, String] :cluster_arn
@@ -23510,6 +23649,87 @@ module Aws::SageMaker
23510
23649
  req.send_request(options)
23511
23650
  end
23512
23651
 
23652
+ # Retrieves a list of operation summaries for a specified HyperPod
23653
+ # cluster, such as scaling operations and deep health checks. The
23654
+ # operation supports filtering, sorting, and pagination of results.
23655
+ #
23656
+ # @option params [required, String] :cluster_name
23657
+ # The name or Amazon Resource Name (ARN) of the HyperPod cluster for
23658
+ # which to list operations.
23659
+ #
23660
+ # @option params [String] :status
23661
+ # The operation status to filter by. If specified, only operations with
23662
+ # this status are returned. Valid values are `InProgress`, `Superseded`,
23663
+ # or `Done`.
23664
+ #
23665
+ # @option params [String] :type
23666
+ # The operation type to filter by. If specified, only operations of this
23667
+ # type are returned. Valid values are `Scaling` or `DeepHealthCheck`.
23668
+ #
23669
+ # @option params [Time,DateTime,Date,Integer,String] :creation_time_after
23670
+ # The start of the time range for filtering operations. Only operations
23671
+ # created after this time are included in the results.
23672
+ #
23673
+ # @option params [Time,DateTime,Date,Integer,String] :creation_time_before
23674
+ # The end of the time range for filtering operations. Only operations
23675
+ # created before this time are included in the results.
23676
+ #
23677
+ # @option params [String] :sort_by
23678
+ # The field to use for sorting the operation list. Currently, the only
23679
+ # supported value is `CreationTime`.
23680
+ #
23681
+ # @option params [String] :sort_order
23682
+ # The order in which to sort the results. Valid values are `Ascending`
23683
+ # or `Descending` (the default is `Descending`).
23684
+ #
23685
+ # @option params [Integer] :max_results
23686
+ # The maximum number of operations to return in the response. Valid
23687
+ # range is 1 to 100.
23688
+ #
23689
+ # @option params [String] :next_token
23690
+ # A token to retrieve the next set of results. This token is obtained
23691
+ # from the output of a previous `ListClusterOperations` call.
23692
+ #
23693
+ # @return [Types::ListClusterOperationsResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
23694
+ #
23695
+ # * {Types::ListClusterOperationsResponse#next_token #next_token} => String
23696
+ # * {Types::ListClusterOperationsResponse#cluster_operation_summaries #cluster_operation_summaries} => Array<Types::ClusterOperationSummary>
23697
+ #
23698
+ # The returned {Seahorse::Client::Response response} is a pageable response and is Enumerable. For details on usage see {Aws::PageableResponse PageableResponse}.
23699
+ #
23700
+ # @example Request syntax with placeholder values
23701
+ #
23702
+ # resp = client.list_cluster_operations({
23703
+ # cluster_name: "ClusterNameOrArn", # required
23704
+ # status: "InProgress", # accepts InProgress, Superseded, Done
23705
+ # type: "Scaling", # accepts Scaling, DeepHealthCheck
23706
+ # creation_time_after: Time.now,
23707
+ # creation_time_before: Time.now,
23708
+ # sort_by: "CreationTime", # accepts CreationTime
23709
+ # sort_order: "Ascending", # accepts Ascending, Descending
23710
+ # max_results: 1,
23711
+ # next_token: "NextToken",
23712
+ # })
23713
+ #
23714
+ # @example Response structure
23715
+ #
23716
+ # resp.next_token #=> String
23717
+ # resp.cluster_operation_summaries #=> Array
23718
+ # resp.cluster_operation_summaries[0].operation_id #=> String
23719
+ # resp.cluster_operation_summaries[0].cluster_arn #=> String
23720
+ # resp.cluster_operation_summaries[0].type #=> String, one of "Scaling", "DeepHealthCheck"
23721
+ # resp.cluster_operation_summaries[0].status #=> String, one of "InProgress", "Superseded", "Done"
23722
+ # resp.cluster_operation_summaries[0].creation_time #=> Time
23723
+ #
23724
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ListClusterOperations AWS API Documentation
23725
+ #
23726
+ # @overload list_cluster_operations(params = {})
23727
+ # @param [Hash] params ({})
23728
+ def list_cluster_operations(params = {}, options = {})
23729
+ req = build_request(:list_cluster_operations, params)
23730
+ req.send_request(options)
23731
+ end
23732
+
23513
23733
  # List the cluster policy configurations.
23514
23734
  #
23515
23735
  # @option params [Time,DateTime,Date,Integer,String] :created_after
@@ -31405,7 +31625,7 @@ module Aws::SageMaker
31405
31625
  # },
31406
31626
  # image_id: "ImageId",
31407
31627
  # auto_patch_config: {
31408
- # patching_strategy: "WhenIdle", # required, accepts WhenIdle, WhenAllIdle
31628
+ # patching_strategy: "WhenIdle", # required, accepts WhenIdle, WhenAllIdle, DrainFirst
31409
31629
  # patch_schedule: {
31410
31630
  # next_patch_date: Time.now,
31411
31631
  # },
@@ -31427,6 +31647,11 @@ module Aws::SageMaker
31427
31647
  # },
31428
31648
  # ],
31429
31649
  # },
31650
+ # idle_wait_window_in_seconds: 1,
31651
+ # drain_policy: {
31652
+ # drain_timeout_in_seconds: 1,
31653
+ # on_drain_timeout: "Force", # accepts Force, Skip
31654
+ # },
31430
31655
  # },
31431
31656
  # image_release_version: "ImageReleaseVersion",
31432
31657
  # kubernetes_config: {
@@ -31580,6 +31805,11 @@ module Aws::SageMaker
31580
31805
  # an earlier version. To use an earlier version, delete the extension
31581
31806
  # and create it again with that version.
31582
31807
  #
31808
+ # To retry the operation on nodes where it failed, fix the cause that
31809
+ # `DescribeClusterExtension` reports in `Health`, and then send an
31810
+ # `UpdateClusterExtension` request that doesn't change the version or
31811
+ # the configuration.
31812
+ #
31583
31813
  # You can't update an extension while it is being deleted.
31584
31814
  #
31585
31815
  # @option params [required, String] :cluster_arn
@@ -35579,7 +35809,7 @@ module Aws::SageMaker
35579
35809
  tracer: tracer
35580
35810
  )
35581
35811
  context[:gem_name] = 'aws-sdk-sagemaker'
35582
- context[:gem_version] = '1.414.0'
35812
+ context[:gem_version] = '1.415.0'
35583
35813
  Seahorse::Client::Request.new(handlers, context)
35584
35814
  end
35585
35815