2025/11/21 - Amazon SageMaker Service - 8 updated api methods
Changes Enhanced SageMaker HyperPod instance groups with support for MinInstanceCount, CapacityRequirements (Spot/On-Demand), and KubernetesConfig (labels and taints). Also Added speculative decoding and MaxInstanceCount for model optimization jobs.
{'InstanceGroups': {'CapacityRequirements': {'OnDemand': {}, 'Spot': {}},
'KubernetesConfig': {'Labels': {'string': 'string'},
'Taints': [{'Effect': 'NoSchedule | '
'PreferNoSchedule '
'| NoExecute',
'Key': 'string',
'Value': 'string'}]},
'MinInstanceCount': 'integer'}}
Creates a SageMaker HyperPod cluster. SageMaker HyperPod is a capability of SageMaker for creating and managing persistent clusters for developing large machine learning models, such as large language models (LLMs) and diffusion models. To learn more, see Amazon SageMaker HyperPod in the Amazon SageMaker Developer Guide.
See also: AWS API Documentation
Request Syntax
client.create_cluster(
ClusterName='string',
InstanceGroups=[
{
'InstanceCount': 123,
'MinInstanceCount': 123,
'InstanceGroupName': 'string',
'InstanceType': 'ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p6e-gb200.36xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.c5.large'|'ml.c5.xlarge'|'ml.c5.2xlarge'|'ml.c5.4xlarge'|'ml.c5.9xlarge'|'ml.c5.12xlarge'|'ml.c5.18xlarge'|'ml.c5.24xlarge'|'ml.c5n.large'|'ml.c5n.2xlarge'|'ml.c5n.4xlarge'|'ml.c5n.9xlarge'|'ml.c5n.18xlarge'|'ml.m5.large'|'ml.m5.xlarge'|'ml.m5.2xlarge'|'ml.m5.4xlarge'|'ml.m5.8xlarge'|'ml.m5.12xlarge'|'ml.m5.16xlarge'|'ml.m5.24xlarge'|'ml.t3.medium'|'ml.t3.large'|'ml.t3.xlarge'|'ml.t3.2xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.16xlarge'|'ml.g6.12xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.gr6.4xlarge'|'ml.gr6.8xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.16xlarge'|'ml.g6e.12xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.p6-b200.48xlarge'|'ml.trn2.3xlarge'|'ml.trn2.48xlarge'|'ml.c6i.large'|'ml.c6i.xlarge'|'ml.c6i.2xlarge'|'ml.c6i.4xlarge'|'ml.c6i.8xlarge'|'ml.c6i.12xlarge'|'ml.c6i.16xlarge'|'ml.c6i.24xlarge'|'ml.c6i.32xlarge'|'ml.m6i.large'|'ml.m6i.xlarge'|'ml.m6i.2xlarge'|'ml.m6i.4xlarge'|'ml.m6i.8xlarge'|'ml.m6i.12xlarge'|'ml.m6i.16xlarge'|'ml.m6i.24xlarge'|'ml.m6i.32xlarge'|'ml.r6i.large'|'ml.r6i.xlarge'|'ml.r6i.2xlarge'|'ml.r6i.4xlarge'|'ml.r6i.8xlarge'|'ml.r6i.12xlarge'|'ml.r6i.16xlarge'|'ml.r6i.24xlarge'|'ml.r6i.32xlarge'|'ml.i3en.large'|'ml.i3en.xlarge'|'ml.i3en.2xlarge'|'ml.i3en.3xlarge'|'ml.i3en.6xlarge'|'ml.i3en.12xlarge'|'ml.i3en.24xlarge'|'ml.m7i.large'|'ml.m7i.xlarge'|'ml.m7i.2xlarge'|'ml.m7i.4xlarge'|'ml.m7i.8xlarge'|'ml.m7i.12xlarge'|'ml.m7i.16xlarge'|'ml.m7i.24xlarge'|'ml.m7i.48xlarge'|'ml.r7i.large'|'ml.r7i.xlarge'|'ml.r7i.2xlarge'|'ml.r7i.4xlarge'|'ml.r7i.8xlarge'|'ml.r7i.12xlarge'|'ml.r7i.16xlarge'|'ml.r7i.24xlarge'|'ml.r7i.48xlarge',
'LifeCycleConfig': {
'SourceS3Uri': 'string',
'OnCreate': 'string'
},
'ExecutionRole': 'string',
'ThreadsPerCore': 123,
'InstanceStorageConfigs': [
{
'EbsVolumeConfig': {
'VolumeSizeInGB': 123,
'VolumeKmsKeyId': 'string',
'RootVolume': True|False
}
},
],
'OnStartDeepHealthChecks': [
'InstanceStress'|'InstanceConnectivity',
],
'TrainingPlanArn': 'string',
'OverrideVpcConfig': {
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
},
'ScheduledUpdateConfig': {
'ScheduleExpression': 'string',
'DeploymentConfig': {
'RollingUpdatePolicy': {
'MaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
},
'RollbackMaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
}
},
'WaitIntervalInSeconds': 123,
'AutoRollbackConfiguration': [
{
'AlarmName': 'string'
},
]
}
},
'ImageId': 'string',
'KubernetesConfig': {
'Labels': {
'string': 'string'
},
'Taints': [
{
'Key': 'string',
'Value': 'string',
'Effect': 'NoSchedule'|'PreferNoSchedule'|'NoExecute'
},
]
},
'CapacityRequirements': {
'Spot': {}
,
'OnDemand': {}
}
},
],
RestrictedInstanceGroups=[
{
'InstanceCount': 123,
'InstanceGroupName': 'string',
'InstanceType': 'ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p6e-gb200.36xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.c5.large'|'ml.c5.xlarge'|'ml.c5.2xlarge'|'ml.c5.4xlarge'|'ml.c5.9xlarge'|'ml.c5.12xlarge'|'ml.c5.18xlarge'|'ml.c5.24xlarge'|'ml.c5n.large'|'ml.c5n.2xlarge'|'ml.c5n.4xlarge'|'ml.c5n.9xlarge'|'ml.c5n.18xlarge'|'ml.m5.large'|'ml.m5.xlarge'|'ml.m5.2xlarge'|'ml.m5.4xlarge'|'ml.m5.8xlarge'|'ml.m5.12xlarge'|'ml.m5.16xlarge'|'ml.m5.24xlarge'|'ml.t3.medium'|'ml.t3.large'|'ml.t3.xlarge'|'ml.t3.2xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.16xlarge'|'ml.g6.12xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.gr6.4xlarge'|'ml.gr6.8xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.16xlarge'|'ml.g6e.12xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.p6-b200.48xlarge'|'ml.trn2.3xlarge'|'ml.trn2.48xlarge'|'ml.c6i.large'|'ml.c6i.xlarge'|'ml.c6i.2xlarge'|'ml.c6i.4xlarge'|'ml.c6i.8xlarge'|'ml.c6i.12xlarge'|'ml.c6i.16xlarge'|'ml.c6i.24xlarge'|'ml.c6i.32xlarge'|'ml.m6i.large'|'ml.m6i.xlarge'|'ml.m6i.2xlarge'|'ml.m6i.4xlarge'|'ml.m6i.8xlarge'|'ml.m6i.12xlarge'|'ml.m6i.16xlarge'|'ml.m6i.24xlarge'|'ml.m6i.32xlarge'|'ml.r6i.large'|'ml.r6i.xlarge'|'ml.r6i.2xlarge'|'ml.r6i.4xlarge'|'ml.r6i.8xlarge'|'ml.r6i.12xlarge'|'ml.r6i.16xlarge'|'ml.r6i.24xlarge'|'ml.r6i.32xlarge'|'ml.i3en.large'|'ml.i3en.xlarge'|'ml.i3en.2xlarge'|'ml.i3en.3xlarge'|'ml.i3en.6xlarge'|'ml.i3en.12xlarge'|'ml.i3en.24xlarge'|'ml.m7i.large'|'ml.m7i.xlarge'|'ml.m7i.2xlarge'|'ml.m7i.4xlarge'|'ml.m7i.8xlarge'|'ml.m7i.12xlarge'|'ml.m7i.16xlarge'|'ml.m7i.24xlarge'|'ml.m7i.48xlarge'|'ml.r7i.large'|'ml.r7i.xlarge'|'ml.r7i.2xlarge'|'ml.r7i.4xlarge'|'ml.r7i.8xlarge'|'ml.r7i.12xlarge'|'ml.r7i.16xlarge'|'ml.r7i.24xlarge'|'ml.r7i.48xlarge',
'ExecutionRole': 'string',
'ThreadsPerCore': 123,
'InstanceStorageConfigs': [
{
'EbsVolumeConfig': {
'VolumeSizeInGB': 123,
'VolumeKmsKeyId': 'string',
'RootVolume': True|False
}
},
],
'OnStartDeepHealthChecks': [
'InstanceStress'|'InstanceConnectivity',
],
'TrainingPlanArn': 'string',
'OverrideVpcConfig': {
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
},
'ScheduledUpdateConfig': {
'ScheduleExpression': 'string',
'DeploymentConfig': {
'RollingUpdatePolicy': {
'MaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
},
'RollbackMaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
}
},
'WaitIntervalInSeconds': 123,
'AutoRollbackConfiguration': [
{
'AlarmName': 'string'
},
]
}
},
'EnvironmentConfig': {
'FSxLustreConfig': {
'SizeInGiB': 123,
'PerUnitStorageThroughput': 123
}
}
},
],
VpcConfig={
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
},
Tags=[
{
'Key': 'string',
'Value': 'string'
},
],
Orchestrator={
'Eks': {
'ClusterArn': 'string'
}
},
NodeRecovery='Automatic'|'None',
TieredStorageConfig={
'Mode': 'Enable'|'Disable',
'InstanceMemoryAllocationPercentage': 123
},
NodeProvisioningMode='Continuous',
ClusterRole='string',
AutoScaling={
'Mode': 'Enable'|'Disable',
'AutoScalerType': 'Karpenter'
}
)
string
[REQUIRED]
The name for the new SageMaker HyperPod cluster.
list
The instance groups to be created in the SageMaker HyperPod cluster.
(dict) --
The specifications of an instance group that you need to define.
InstanceCount (integer) -- [REQUIRED]
Specifies the number of instances to add to the instance group of a SageMaker HyperPod cluster.
MinInstanceCount (integer) --
Defines the minimum number of instances required for an instance group to become InService. If this threshold isn't met within 3 hours, the instance group rolls back to its previous state - zero instances for new instance groups, or previous settings for existing instance groups. MinInstanceCount only affects the initial transition to InService and does not guarantee maintaining this minimum afterward.
InstanceGroupName (string) -- [REQUIRED]
Specifies the name of the instance group.
InstanceType (string) -- [REQUIRED]
Specifies the instance type of the instance group.
LifeCycleConfig (dict) -- [REQUIRED]
Specifies the LifeCycle configuration for the instance group.
SourceS3Uri (string) -- [REQUIRED]
An Amazon S3 bucket path where your lifecycle scripts are stored.
OnCreate (string) -- [REQUIRED]
The file name of the entrypoint script of lifecycle scripts under SourceS3Uri. This entrypoint script runs during cluster creation.
ExecutionRole (string) -- [REQUIRED]
Specifies an IAM execution role to be assumed by the instance group.
ThreadsPerCore (integer) --
Specifies the value for Threads per core. For instance types that support multithreading, you can specify 1 for disabling multithreading and 2 for enabling multithreading. For instance types that doesn't support multithreading, specify 1. For more information, see the reference table of CPU cores and threads per CPU core per instance type in the Amazon Elastic Compute Cloud User Guide.
InstanceStorageConfigs (list) --
Specifies the additional storage configurations for the instances in the SageMaker HyperPod cluster instance group.
(dict) --
Defines the configuration for attaching additional storage to the instances in the SageMaker HyperPod cluster instance group. To learn more, see SageMaker HyperPod release notes: June 20, 2024.
EbsVolumeConfig (dict) --
Defines the configuration for attaching additional Amazon Elastic Block Store (EBS) volumes to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeSizeInGB (integer) --
The size in gigabytes (GB) of the additional EBS volume to be attached to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeKmsKeyId (string) --
The ID of a KMS key to encrypt the Amazon EBS volume.
RootVolume (boolean) --
Specifies whether the configuration is for the cluster's root or secondary Amazon EBS volume. You can specify two ClusterEbsVolumeConfig fields to configure both the root and secondary volumes. Set the value to True if you'd like to provide your own customer managed Amazon Web Services KMS key to encrypt the root volume. When True:
The configuration is applied to the root volume.
You can't specify the VolumeSizeInGB field. The size of the root volume is determined for you.
You must specify a KMS key ID for VolumeKmsKeyId to encrypt the root volume with your own KMS key instead of an Amazon Web Services owned KMS key.
Otherwise, by default, the value is False, and the following applies:
The configuration is applied to the secondary volume, while the root volume is encrypted with an Amazon Web Services owned key.
You must specify the VolumeSizeInGB field.
You can optionally specify the VolumeKmsKeyId to encrypt the secondary volume with your own KMS key instead of an Amazon Web Services owned KMS key.
OnStartDeepHealthChecks (list) --
A flag indicating whether deep health checks should be performed when the cluster instance group is created or updated.
(string) --
TrainingPlanArn (string) --
The Amazon Resource Name (ARN); of the training plan to use for this cluster instance group.
For more information about how to reserve GPU capacity for your SageMaker HyperPod clusters using Amazon SageMaker Training Plan, see ``CreateTrainingPlan ``.
OverrideVpcConfig (dict) --
To configure multi-AZ deployments, customize the Amazon VPC configuration at the instance group level. You can specify different subnets and security groups across different AZs in the instance group specification to override a SageMaker HyperPod cluster's default Amazon VPC configuration. For more information about deploying a cluster in multiple AZs, see Setting up SageMaker HyperPod clusters across multiple AZs.
SecurityGroupIds (list) -- [REQUIRED]
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) -- [REQUIRED]
The ID of the subnets in the VPC to which you want to connect your training job or model. For information about the availability of specific instance types, see Supported Instance Types and Availability Zones.
(string) --
ScheduledUpdateConfig (dict) --
The configuration object of the schedule that SageMaker uses to update the AMI.
ScheduleExpression (string) -- [REQUIRED]
A cron expression that specifies the schedule that SageMaker follows when updating the AMI.
DeploymentConfig (dict) --
The configuration to use when updating the AMI versions.
RollingUpdatePolicy (dict) --
The policy that SageMaker uses when updating the AMI versions of the cluster.
MaximumBatchSize (dict) -- [REQUIRED]
The maximum amount of instances in the cluster that SageMaker can update at a time.
Type (string) -- [REQUIRED]
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) -- [REQUIRED]
Specifies the amount or percentage of instances SageMaker updates at a time.
RollbackMaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can roll back at a time.
Type (string) -- [REQUIRED]
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) -- [REQUIRED]
Specifies the amount or percentage of instances SageMaker updates at a time.
WaitIntervalInSeconds (integer) --
The duration in seconds that SageMaker waits before updating more instances in the cluster.
AutoRollbackConfiguration (list) --
An array that contains the alarms that SageMaker monitors to know whether to roll back the AMI update.
(dict) --
The details of the alarm to monitor during the AMI update.
AlarmName (string) -- [REQUIRED]
The name of the alarm.
ImageId (string) --
When configuring your HyperPod cluster, you can specify an image ID using one of the following options:
HyperPodPublicAmiId: Use a HyperPod public AMI
CustomAmiId: Use your custom AMI
default: Use the default latest system image
If you choose to use a custom AMI ( CustomAmiId), ensure it meets the following requirements:
Encryption: The custom AMI must be unencrypted.
Ownership: The custom AMI must be owned by the same Amazon Web Services account that is creating the HyperPod cluster.
Volume support: Only the primary AMI snapshot volume is supported; additional AMI volumes are not supported.
When updating the instance group's AMI through the UpdateClusterSoftware operation, if an instance group uses a custom AMI, you must provide an ImageId or use the default as input. Note that if you don't specify an instance group in your UpdateClusterSoftware request, then all of the instance groups are patched with the specified image.
KubernetesConfig (dict) --
Specifies the Kubernetes configuration for the instance group. You describe what you want the labels and taints to look like, and the cluster works to reconcile the actual state with the declared state for nodes in this instance group.
Labels (dict) --
Key-value pairs of labels to be applied to cluster nodes.
(string) --
(string) --
Taints (list) --
List of taints to be applied to cluster nodes.
(dict) --
A Kubernetes taint that can be applied to cluster nodes.
Key (string) -- [REQUIRED]
The key of the taint.
Value (string) --
The value of the taint.
Effect (string) -- [REQUIRED]
The effect of the taint. Valid values are NoSchedule, PreferNoSchedule, and NoExecute.
CapacityRequirements (dict) --
Specifies the capacity requirements for the instance group.
Spot (dict) --
Configuration options specific to Spot instances.
OnDemand (dict) --
Configuration options specific to On-Demand instances.
list
The specialized instance groups for training models like Amazon Nova to be created in the SageMaker HyperPod cluster.
(dict) --
The specifications of a restricted instance group that you need to define.
InstanceCount (integer) -- [REQUIRED]
Specifies the number of instances to add to the restricted instance group of a SageMaker HyperPod cluster.
InstanceGroupName (string) -- [REQUIRED]
Specifies the name of the restricted instance group.
InstanceType (string) -- [REQUIRED]
Specifies the instance type of the restricted instance group.
ExecutionRole (string) -- [REQUIRED]
Specifies an IAM execution role to be assumed by the restricted instance group.
ThreadsPerCore (integer) --
The number you specified to TreadsPerCore in CreateCluster for enabling or disabling multithreading. For instance types that support multithreading, you can specify 1 for disabling multithreading and 2 for enabling multithreading. For more information, see the reference table of CPU cores and threads per CPU core per instance type in the Amazon Elastic Compute Cloud User Guide.
InstanceStorageConfigs (list) --
Specifies the additional storage configurations for the instances in the SageMaker HyperPod cluster restricted instance group.
(dict) --
Defines the configuration for attaching additional storage to the instances in the SageMaker HyperPod cluster instance group. To learn more, see SageMaker HyperPod release notes: June 20, 2024.
EbsVolumeConfig (dict) --
Defines the configuration for attaching additional Amazon Elastic Block Store (EBS) volumes to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeSizeInGB (integer) --
The size in gigabytes (GB) of the additional EBS volume to be attached to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeKmsKeyId (string) --
The ID of a KMS key to encrypt the Amazon EBS volume.
RootVolume (boolean) --
Specifies whether the configuration is for the cluster's root or secondary Amazon EBS volume. You can specify two ClusterEbsVolumeConfig fields to configure both the root and secondary volumes. Set the value to True if you'd like to provide your own customer managed Amazon Web Services KMS key to encrypt the root volume. When True:
The configuration is applied to the root volume.
You can't specify the VolumeSizeInGB field. The size of the root volume is determined for you.
You must specify a KMS key ID for VolumeKmsKeyId to encrypt the root volume with your own KMS key instead of an Amazon Web Services owned KMS key.
Otherwise, by default, the value is False, and the following applies:
The configuration is applied to the secondary volume, while the root volume is encrypted with an Amazon Web Services owned key.
You must specify the VolumeSizeInGB field.
You can optionally specify the VolumeKmsKeyId to encrypt the secondary volume with your own KMS key instead of an Amazon Web Services owned KMS key.
OnStartDeepHealthChecks (list) --
A flag indicating whether deep health checks should be performed when the cluster restricted instance group is created or updated.
(string) --
TrainingPlanArn (string) --
The Amazon Resource Name (ARN) of the training plan to filter clusters by. For more information about reserving GPU capacity for your SageMaker HyperPod clusters using Amazon SageMaker Training Plan, see ``CreateTrainingPlan ``.
OverrideVpcConfig (dict) --
Specifies an Amazon Virtual Private Cloud (VPC) that your SageMaker jobs, hosted models, and compute resources have access to. You can control access to and from your resources by configuring a VPC. For more information, see Give SageMaker Access to Resources in your Amazon VPC.
SecurityGroupIds (list) -- [REQUIRED]
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) -- [REQUIRED]
The ID of the subnets in the VPC to which you want to connect your training job or model. For information about the availability of specific instance types, see Supported Instance Types and Availability Zones.
(string) --
ScheduledUpdateConfig (dict) --
The configuration object of the schedule that SageMaker follows when updating the AMI.
ScheduleExpression (string) -- [REQUIRED]
A cron expression that specifies the schedule that SageMaker follows when updating the AMI.
DeploymentConfig (dict) --
The configuration to use when updating the AMI versions.
RollingUpdatePolicy (dict) --
The policy that SageMaker uses when updating the AMI versions of the cluster.
MaximumBatchSize (dict) -- [REQUIRED]
The maximum amount of instances in the cluster that SageMaker can update at a time.
Type (string) -- [REQUIRED]
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) -- [REQUIRED]
Specifies the amount or percentage of instances SageMaker updates at a time.
RollbackMaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can roll back at a time.
Type (string) -- [REQUIRED]
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) -- [REQUIRED]
Specifies the amount or percentage of instances SageMaker updates at a time.
WaitIntervalInSeconds (integer) --
The duration in seconds that SageMaker waits before updating more instances in the cluster.
AutoRollbackConfiguration (list) --
An array that contains the alarms that SageMaker monitors to know whether to roll back the AMI update.
(dict) --
The details of the alarm to monitor during the AMI update.
AlarmName (string) -- [REQUIRED]
The name of the alarm.
EnvironmentConfig (dict) -- [REQUIRED]
The configuration for the restricted instance groups (RIG) environment.
FSxLustreConfig (dict) --
Configuration settings for an Amazon FSx for Lustre file system to be used with the cluster.
SizeInGiB (integer) -- [REQUIRED]
The storage capacity of the Amazon FSx for Lustre file system, specified in gibibytes (GiB).
PerUnitStorageThroughput (integer) -- [REQUIRED]
The throughput capacity of the Amazon FSx for Lustre file system, measured in MB/s per TiB of storage.
dict
Specifies the Amazon Virtual Private Cloud (VPC) that is associated with the Amazon SageMaker HyperPod cluster. You can control access to and from your resources by configuring your VPC. For more information, see Give SageMaker access to resources in your Amazon VPC.
SecurityGroupIds (list) -- [REQUIRED]
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) -- [REQUIRED]
The ID of the subnets in the VPC to which you want to connect your training job or model. For information about the availability of specific instance types, see Supported Instance Types and Availability Zones.
(string) --
list
Custom tags for managing the SageMaker HyperPod cluster as an Amazon Web Services resource. You can add tags to your cluster in the same way you add them in other Amazon Web Services services that support tagging. To learn more about tagging Amazon Web Services resources in general, see Tagging Amazon Web Services Resources User Guide.
(dict) --
A tag object that consists of a key and an optional value, used to manage metadata for SageMaker Amazon Web Services resources.
You can add tags to notebook instances, training jobs, hyperparameter tuning jobs, batch transform jobs, models, labeling jobs, work teams, endpoint configurations, and endpoints. For more information on adding tags to SageMaker resources, see AddTags.
For more information on adding metadata to your Amazon Web Services resources with tagging, see Tagging Amazon Web Services resources. For advice on best practices for managing Amazon Web Services resources with tagging, see Tagging Best Practices: Implement an Effective Amazon Web Services Resource Tagging Strategy.
Key (string) -- [REQUIRED]
The tag key. Tag keys must be unique per resource.
Value (string) -- [REQUIRED]
The tag value.
dict
The type of orchestrator to use for the SageMaker HyperPod cluster. Currently, the only supported value is "eks", which is to use an Amazon Elastic Kubernetes Service cluster as the orchestrator.
Eks (dict) -- [REQUIRED]
The Amazon EKS cluster used as the orchestrator for the SageMaker HyperPod cluster.
ClusterArn (string) -- [REQUIRED]
The Amazon Resource Name (ARN) of the Amazon EKS cluster associated with the SageMaker HyperPod cluster.
string
The node recovery mode for the SageMaker HyperPod cluster. When set to Automatic, SageMaker HyperPod will automatically reboot or replace faulty nodes when issues are detected. When set to None, cluster administrators will need to manually manage any faulty cluster instances.
dict
The configuration for managed tier checkpointing on the HyperPod cluster. When enabled, this feature uses a multi-tier storage approach for storing model checkpoints, providing faster checkpoint operations and improved fault tolerance across cluster nodes.
Mode (string) -- [REQUIRED]
Specifies whether managed tier checkpointing is enabled or disabled for the HyperPod cluster. When set to Enable, the system installs a memory management daemon that provides disaggregated memory as a service for checkpoint storage. When set to Disable, the feature is turned off and the memory management daemon is removed from the cluster.
InstanceMemoryAllocationPercentage (integer) --
The percentage (int) of cluster memory to allocate for checkpointing.
string
The mode for provisioning nodes in the cluster. You can specify the following modes:
Continuous: Scaling behavior that enables 1) concurrent operation execution within instance groups, 2) continuous retry mechanisms for failed operations, 3) enhanced customer visibility into cluster events through detailed event streams, 4) partial provisioning capabilities. Your clusters and instance groups remain InService while scaling. This mode is only supported for EKS orchestrated clusters.
string
The Amazon Resource Name (ARN) of the IAM role that HyperPod assumes to perform cluster autoscaling operations. This role must have permissions for sagemaker:BatchAddClusterNodes and sagemaker:BatchDeleteClusterNodes. This is only required when autoscaling is enabled and when HyperPod is performing autoscaling operations.
dict
The autoscaling configuration for the cluster. Enables automatic scaling of cluster nodes based on workload demand using a Karpenter-based system.
Mode (string) -- [REQUIRED]
Describes whether autoscaling is enabled or disabled for the cluster. Valid values are Enable and Disable.
AutoScalerType (string) --
The type of autoscaler to use. Currently supported value is Karpenter.
dict
Response Syntax
{
'ClusterArn': 'string'
}
Response Structure
(dict) --
ClusterArn (string) --
The Amazon Resource Name (ARN) of the cluster.
{'DeploymentInstanceType': {'ml.p5en.48xlarge', 'ml.p5e.48xlarge'},
'MaxInstanceCount': 'integer',
'ModelSource': {'SageMakerModel': {'ModelName': 'string'}},
'OptimizationConfigs': {'ModelSpeculativeDecodingConfig': {'Technique': 'EAGLE',
'TrainingDataSource': {'S3DataType': 'S3Prefix '
'| '
'ManifestFile',
'S3Uri': 'string'}}},
'OutputConfig': {'SageMakerModel': {'ModelName': 'string'}}}
Creates a job that optimizes a model for inference performance. To create the job, you provide the location of a source model, and you provide the settings for the optimization techniques that you want the job to apply. When the job completes successfully, SageMaker uploads the new optimized model to the output destination that you specify.
For more information about how to use this action, and about the supported optimization techniques, see Optimize model inference with Amazon SageMaker.
See also: AWS API Documentation
Request Syntax
client.create_optimization_job(
OptimizationJobName='string',
RoleArn='string',
ModelSource={
'S3': {
'S3Uri': 'string',
'ModelAccessConfig': {
'AcceptEula': True|False
}
},
'SageMakerModel': {
'ModelName': 'string'
}
},
DeploymentInstanceType='ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.12xlarge'|'ml.g6.16xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.12xlarge'|'ml.g6e.16xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.inf2.xlarge'|'ml.inf2.8xlarge'|'ml.inf2.24xlarge'|'ml.inf2.48xlarge'|'ml.trn1.2xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge',
MaxInstanceCount=123,
OptimizationEnvironment={
'string': 'string'
},
OptimizationConfigs=[
{
'ModelQuantizationConfig': {
'Image': 'string',
'OverrideEnvironment': {
'string': 'string'
}
},
'ModelCompilationConfig': {
'Image': 'string',
'OverrideEnvironment': {
'string': 'string'
}
},
'ModelShardingConfig': {
'Image': 'string',
'OverrideEnvironment': {
'string': 'string'
}
},
'ModelSpeculativeDecodingConfig': {
'Technique': 'EAGLE',
'TrainingDataSource': {
'S3Uri': 'string',
'S3DataType': 'S3Prefix'|'ManifestFile'
}
}
},
],
OutputConfig={
'KmsKeyId': 'string',
'S3OutputLocation': 'string',
'SageMakerModel': {
'ModelName': 'string'
}
},
StoppingCondition={
'MaxRuntimeInSeconds': 123,
'MaxWaitTimeInSeconds': 123,
'MaxPendingTimeInSeconds': 123
},
Tags=[
{
'Key': 'string',
'Value': 'string'
},
],
VpcConfig={
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
}
)
string
[REQUIRED]
A custom name for the new optimization job.
string
[REQUIRED]
The Amazon Resource Name (ARN) of an IAM role that enables Amazon SageMaker AI to perform tasks on your behalf.
During model optimization, Amazon SageMaker AI needs your permission to:
Read input data from an S3 bucket
Write model artifacts to an S3 bucket
Write logs to Amazon CloudWatch Logs
Publish metrics to Amazon CloudWatch
You grant permissions for all of these tasks to an IAM role. To pass this role to Amazon SageMaker AI, the caller of this API must have the iam:PassRole permission. For more information, see Amazon SageMaker AI Roles.
dict
[REQUIRED]
The location of the source model to optimize with an optimization job.
S3 (dict) --
The Amazon S3 location of a source model to optimize with an optimization job.
S3Uri (string) --
An Amazon S3 URI that locates a source model to optimize with an optimization job.
ModelAccessConfig (dict) --
The access configuration settings for the source ML model for an optimization job, where you can accept the model end-user license agreement (EULA).
AcceptEula (boolean) -- [REQUIRED]
Specifies agreement to the model end-user license agreement (EULA). The AcceptEula value must be explicitly defined as True in order to accept the EULA that this model requires. You are responsible for reviewing and complying with any applicable license terms and making sure they are acceptable for your use case before downloading or using a model.
SageMakerModel (dict) --
The name of an existing SageMaker model to optimize with an optimization job.
ModelName (string) --
The name of a SageMaker model.
string
[REQUIRED]
The type of instance that hosts the optimized model that you create with the optimization job.
integer
The maximum number of instances to use for the optimization job.
dict
The environment variables to set in the model container.
(string) --
(string) --
list
[REQUIRED]
Settings for each of the optimization techniques that the job applies.
(dict) --
Settings for an optimization technique that you apply with a model optimization job.
ModelQuantizationConfig (dict) --
Settings for the model quantization technique that's applied by a model optimization job.
Image (string) --
The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.
OverrideEnvironment (dict) --
Environment variables that override the default ones in the model container.
(string) --
(string) --
ModelCompilationConfig (dict) --
Settings for the model compilation technique that's applied by a model optimization job.
Image (string) --
The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.
OverrideEnvironment (dict) --
Environment variables that override the default ones in the model container.
(string) --
(string) --
ModelShardingConfig (dict) --
Settings for the model sharding technique that's applied by a model optimization job.
Image (string) --
The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.
OverrideEnvironment (dict) --
Environment variables that override the default ones in the model container.
(string) --
(string) --
ModelSpeculativeDecodingConfig (dict) --
Settings for the model speculative decoding technique that's applied by a model optimization job.
Technique (string) -- [REQUIRED]
The speculative decoding technique to apply during model optimization.
TrainingDataSource (dict) --
The location of the training data to use for speculative decoding. The data must be formatted as ShareGPT, OpenAI Completions or OpenAI Chat Completions. The input can also be unencrypted captured data from a SageMaker endpoint as long as the endpoint uses one of the above formats.
S3Uri (string) -- [REQUIRED]
The Amazon S3 URI that points to the training data for speculative decoding.
S3DataType (string) -- [REQUIRED]
The type of data stored in the Amazon S3 location. Valid values are S3Prefix or ManifestFile.
dict
[REQUIRED]
Details for where to store the optimized model that you create with the optimization job.
KmsKeyId (string) --
The Amazon Resource Name (ARN) of a key in Amazon Web Services KMS. SageMaker uses they key to encrypt the artifacts of the optimized model when SageMaker uploads the model to Amazon S3.
S3OutputLocation (string) -- [REQUIRED]
The Amazon S3 URI for where to store the optimized model that you create with an optimization job.
SageMakerModel (dict) --
The name of a SageMaker model to use as the output destination for an optimization job.
ModelName (string) --
The name of a SageMaker model.
dict
[REQUIRED]
Specifies a limit to how long a job can run. When the job reaches the time limit, SageMaker ends the job. Use this API to cap costs.
To stop a training job, SageMaker sends the algorithm the SIGTERM signal, which delays job termination for 120 seconds. Algorithms can use this 120-second window to save the model artifacts, so the results of training are not lost.
The training algorithms provided by SageMaker automatically save the intermediate results of a model training job when possible. This attempt to save artifacts is only a best effort case as model might not be in a state from which it can be saved. For example, if training has just started, the model might not be ready to save. When saved, this intermediate data is a valid model artifact. You can use it to create a model with CreateModel.
MaxRuntimeInSeconds (integer) --
The maximum length of time, in seconds, that a training or compilation job can run before it is stopped.
For compilation jobs, if the job does not complete during this time, a TimeOut error is generated. We recommend starting with 900 seconds and increasing as necessary based on your model.
For all other jobs, if the job does not complete during this time, SageMaker ends the job. When RetryStrategy is specified in the job request, MaxRuntimeInSeconds specifies the maximum time for all of the attempts in total, not each individual attempt. The default value is 1 day. The maximum value is 28 days.
The maximum time that a TrainingJob can run in total, including any time spent publishing metrics or archiving and uploading models after it has been stopped, is 30 days.
MaxWaitTimeInSeconds (integer) --
The maximum length of time, in seconds, that a managed Spot training job has to complete. It is the amount of time spent waiting for Spot capacity plus the amount of time the job can run. It must be equal to or greater than MaxRuntimeInSeconds. If the job does not complete during this time, SageMaker ends the job.
When RetryStrategy is specified in the job request, MaxWaitTimeInSeconds specifies the maximum time for all of the attempts in total, not each individual attempt.
MaxPendingTimeInSeconds (integer) --
The maximum length of time, in seconds, that a training or compilation job can be pending before it is stopped.
list
A list of key-value pairs associated with the optimization job. For more information, see Tagging Amazon Web Services resources in the Amazon Web Services General Reference Guide.
(dict) --
A tag object that consists of a key and an optional value, used to manage metadata for SageMaker Amazon Web Services resources.
You can add tags to notebook instances, training jobs, hyperparameter tuning jobs, batch transform jobs, models, labeling jobs, work teams, endpoint configurations, and endpoints. For more information on adding tags to SageMaker resources, see AddTags.
For more information on adding metadata to your Amazon Web Services resources with tagging, see Tagging Amazon Web Services resources. For advice on best practices for managing Amazon Web Services resources with tagging, see Tagging Best Practices: Implement an Effective Amazon Web Services Resource Tagging Strategy.
Key (string) -- [REQUIRED]
The tag key. Tag keys must be unique per resource.
Value (string) -- [REQUIRED]
The tag value.
dict
A VPC in Amazon VPC that your optimized model has access to.
SecurityGroupIds (list) -- [REQUIRED]
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) -- [REQUIRED]
The ID of the subnets in the VPC to which you want to connect your optimized model.
(string) --
dict
Response Syntax
{
'OptimizationJobArn': 'string'
}
Response Structure
(dict) --
OptimizationJobArn (string) --
The Amazon Resource Name (ARN) of the optimization job.
{'InstanceGroups': {'ActiveOperations': {'Scaling': 'integer'},
'CapacityRequirements': {'OnDemand': {}, 'Spot': {}},
'KubernetesConfig': {'CurrentLabels': {'string': 'string'},
'CurrentTaints': [{'Effect': 'NoSchedule '
'| '
'PreferNoSchedule '
'| '
'NoExecute',
'Key': 'string',
'Value': 'string'}],
'DesiredLabels': {'string': 'string'},
'DesiredTaints': [{'Effect': 'NoSchedule '
'| '
'PreferNoSchedule '
'| '
'NoExecute',
'Key': 'string',
'Value': 'string'}]},
'MinCount': 'integer'}}
Retrieves information of a SageMaker HyperPod cluster.
See also: AWS API Documentation
Request Syntax
client.describe_cluster(
ClusterName='string'
)
string
[REQUIRED]
The string name or the Amazon Resource Name (ARN) of the SageMaker HyperPod cluster.
dict
Response Syntax
{
'ClusterArn': 'string',
'ClusterName': 'string',
'ClusterStatus': 'Creating'|'Deleting'|'Failed'|'InService'|'RollingBack'|'SystemUpdating'|'Updating',
'CreationTime': datetime(2015, 1, 1),
'FailureMessage': 'string',
'InstanceGroups': [
{
'CurrentCount': 123,
'TargetCount': 123,
'MinCount': 123,
'InstanceGroupName': 'string',
'InstanceType': 'ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p6e-gb200.36xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.c5.large'|'ml.c5.xlarge'|'ml.c5.2xlarge'|'ml.c5.4xlarge'|'ml.c5.9xlarge'|'ml.c5.12xlarge'|'ml.c5.18xlarge'|'ml.c5.24xlarge'|'ml.c5n.large'|'ml.c5n.2xlarge'|'ml.c5n.4xlarge'|'ml.c5n.9xlarge'|'ml.c5n.18xlarge'|'ml.m5.large'|'ml.m5.xlarge'|'ml.m5.2xlarge'|'ml.m5.4xlarge'|'ml.m5.8xlarge'|'ml.m5.12xlarge'|'ml.m5.16xlarge'|'ml.m5.24xlarge'|'ml.t3.medium'|'ml.t3.large'|'ml.t3.xlarge'|'ml.t3.2xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.16xlarge'|'ml.g6.12xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.gr6.4xlarge'|'ml.gr6.8xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.16xlarge'|'ml.g6e.12xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.p6-b200.48xlarge'|'ml.trn2.3xlarge'|'ml.trn2.48xlarge'|'ml.c6i.large'|'ml.c6i.xlarge'|'ml.c6i.2xlarge'|'ml.c6i.4xlarge'|'ml.c6i.8xlarge'|'ml.c6i.12xlarge'|'ml.c6i.16xlarge'|'ml.c6i.24xlarge'|'ml.c6i.32xlarge'|'ml.m6i.large'|'ml.m6i.xlarge'|'ml.m6i.2xlarge'|'ml.m6i.4xlarge'|'ml.m6i.8xlarge'|'ml.m6i.12xlarge'|'ml.m6i.16xlarge'|'ml.m6i.24xlarge'|'ml.m6i.32xlarge'|'ml.r6i.large'|'ml.r6i.xlarge'|'ml.r6i.2xlarge'|'ml.r6i.4xlarge'|'ml.r6i.8xlarge'|'ml.r6i.12xlarge'|'ml.r6i.16xlarge'|'ml.r6i.24xlarge'|'ml.r6i.32xlarge'|'ml.i3en.large'|'ml.i3en.xlarge'|'ml.i3en.2xlarge'|'ml.i3en.3xlarge'|'ml.i3en.6xlarge'|'ml.i3en.12xlarge'|'ml.i3en.24xlarge'|'ml.m7i.large'|'ml.m7i.xlarge'|'ml.m7i.2xlarge'|'ml.m7i.4xlarge'|'ml.m7i.8xlarge'|'ml.m7i.12xlarge'|'ml.m7i.16xlarge'|'ml.m7i.24xlarge'|'ml.m7i.48xlarge'|'ml.r7i.large'|'ml.r7i.xlarge'|'ml.r7i.2xlarge'|'ml.r7i.4xlarge'|'ml.r7i.8xlarge'|'ml.r7i.12xlarge'|'ml.r7i.16xlarge'|'ml.r7i.24xlarge'|'ml.r7i.48xlarge',
'LifeCycleConfig': {
'SourceS3Uri': 'string',
'OnCreate': 'string'
},
'ExecutionRole': 'string',
'ThreadsPerCore': 123,
'InstanceStorageConfigs': [
{
'EbsVolumeConfig': {
'VolumeSizeInGB': 123,
'VolumeKmsKeyId': 'string',
'RootVolume': True|False
}
},
],
'OnStartDeepHealthChecks': [
'InstanceStress'|'InstanceConnectivity',
],
'Status': 'InService'|'Creating'|'Updating'|'Failed'|'Degraded'|'SystemUpdating'|'Deleting',
'TrainingPlanArn': 'string',
'TrainingPlanStatus': 'string',
'OverrideVpcConfig': {
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
},
'ScheduledUpdateConfig': {
'ScheduleExpression': 'string',
'DeploymentConfig': {
'RollingUpdatePolicy': {
'MaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
},
'RollbackMaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
}
},
'WaitIntervalInSeconds': 123,
'AutoRollbackConfiguration': [
{
'AlarmName': 'string'
},
]
}
},
'CurrentImageId': 'string',
'DesiredImageId': 'string',
'ActiveOperations': {
'string': 123
},
'KubernetesConfig': {
'CurrentLabels': {
'string': 'string'
},
'DesiredLabels': {
'string': 'string'
},
'CurrentTaints': [
{
'Key': 'string',
'Value': 'string',
'Effect': 'NoSchedule'|'PreferNoSchedule'|'NoExecute'
},
],
'DesiredTaints': [
{
'Key': 'string',
'Value': 'string',
'Effect': 'NoSchedule'|'PreferNoSchedule'|'NoExecute'
},
]
},
'CapacityRequirements': {
'Spot': {},
'OnDemand': {}
},
'TargetStateCount': 123,
'SoftwareUpdateStatus': 'Pending'|'InProgress'|'Succeeded'|'Failed'|'RollbackInProgress'|'RollbackComplete',
'ActiveSoftwareUpdateConfig': {
'RollingUpdatePolicy': {
'MaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
},
'RollbackMaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
}
},
'WaitIntervalInSeconds': 123,
'AutoRollbackConfiguration': [
{
'AlarmName': 'string'
},
]
}
},
],
'RestrictedInstanceGroups': [
{
'CurrentCount': 123,
'TargetCount': 123,
'InstanceGroupName': 'string',
'InstanceType': 'ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p6e-gb200.36xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.c5.large'|'ml.c5.xlarge'|'ml.c5.2xlarge'|'ml.c5.4xlarge'|'ml.c5.9xlarge'|'ml.c5.12xlarge'|'ml.c5.18xlarge'|'ml.c5.24xlarge'|'ml.c5n.large'|'ml.c5n.2xlarge'|'ml.c5n.4xlarge'|'ml.c5n.9xlarge'|'ml.c5n.18xlarge'|'ml.m5.large'|'ml.m5.xlarge'|'ml.m5.2xlarge'|'ml.m5.4xlarge'|'ml.m5.8xlarge'|'ml.m5.12xlarge'|'ml.m5.16xlarge'|'ml.m5.24xlarge'|'ml.t3.medium'|'ml.t3.large'|'ml.t3.xlarge'|'ml.t3.2xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.16xlarge'|'ml.g6.12xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.gr6.4xlarge'|'ml.gr6.8xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.16xlarge'|'ml.g6e.12xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.p6-b200.48xlarge'|'ml.trn2.3xlarge'|'ml.trn2.48xlarge'|'ml.c6i.large'|'ml.c6i.xlarge'|'ml.c6i.2xlarge'|'ml.c6i.4xlarge'|'ml.c6i.8xlarge'|'ml.c6i.12xlarge'|'ml.c6i.16xlarge'|'ml.c6i.24xlarge'|'ml.c6i.32xlarge'|'ml.m6i.large'|'ml.m6i.xlarge'|'ml.m6i.2xlarge'|'ml.m6i.4xlarge'|'ml.m6i.8xlarge'|'ml.m6i.12xlarge'|'ml.m6i.16xlarge'|'ml.m6i.24xlarge'|'ml.m6i.32xlarge'|'ml.r6i.large'|'ml.r6i.xlarge'|'ml.r6i.2xlarge'|'ml.r6i.4xlarge'|'ml.r6i.8xlarge'|'ml.r6i.12xlarge'|'ml.r6i.16xlarge'|'ml.r6i.24xlarge'|'ml.r6i.32xlarge'|'ml.i3en.large'|'ml.i3en.xlarge'|'ml.i3en.2xlarge'|'ml.i3en.3xlarge'|'ml.i3en.6xlarge'|'ml.i3en.12xlarge'|'ml.i3en.24xlarge'|'ml.m7i.large'|'ml.m7i.xlarge'|'ml.m7i.2xlarge'|'ml.m7i.4xlarge'|'ml.m7i.8xlarge'|'ml.m7i.12xlarge'|'ml.m7i.16xlarge'|'ml.m7i.24xlarge'|'ml.m7i.48xlarge'|'ml.r7i.large'|'ml.r7i.xlarge'|'ml.r7i.2xlarge'|'ml.r7i.4xlarge'|'ml.r7i.8xlarge'|'ml.r7i.12xlarge'|'ml.r7i.16xlarge'|'ml.r7i.24xlarge'|'ml.r7i.48xlarge',
'ExecutionRole': 'string',
'ThreadsPerCore': 123,
'InstanceStorageConfigs': [
{
'EbsVolumeConfig': {
'VolumeSizeInGB': 123,
'VolumeKmsKeyId': 'string',
'RootVolume': True|False
}
},
],
'OnStartDeepHealthChecks': [
'InstanceStress'|'InstanceConnectivity',
],
'Status': 'InService'|'Creating'|'Updating'|'Failed'|'Degraded'|'SystemUpdating'|'Deleting',
'TrainingPlanArn': 'string',
'TrainingPlanStatus': 'string',
'OverrideVpcConfig': {
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
},
'ScheduledUpdateConfig': {
'ScheduleExpression': 'string',
'DeploymentConfig': {
'RollingUpdatePolicy': {
'MaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
},
'RollbackMaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
}
},
'WaitIntervalInSeconds': 123,
'AutoRollbackConfiguration': [
{
'AlarmName': 'string'
},
]
}
},
'EnvironmentConfig': {
'FSxLustreConfig': {
'SizeInGiB': 123,
'PerUnitStorageThroughput': 123
},
'S3OutputPath': 'string'
}
},
],
'VpcConfig': {
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
},
'Orchestrator': {
'Eks': {
'ClusterArn': 'string'
}
},
'TieredStorageConfig': {
'Mode': 'Enable'|'Disable',
'InstanceMemoryAllocationPercentage': 123
},
'NodeRecovery': 'Automatic'|'None',
'NodeProvisioningMode': 'Continuous',
'ClusterRole': 'string',
'AutoScaling': {
'Mode': 'Enable'|'Disable',
'AutoScalerType': 'Karpenter',
'Status': 'InService'|'Failed'|'Creating'|'Deleting',
'FailureMessage': 'string'
}
}
Response Structure
(dict) --
ClusterArn (string) --
The Amazon Resource Name (ARN) of the SageMaker HyperPod cluster.
ClusterName (string) --
The name of the SageMaker HyperPod cluster.
ClusterStatus (string) --
The status of the SageMaker HyperPod cluster.
CreationTime (datetime) --
The time when the SageMaker Cluster is created.
FailureMessage (string) --
The failure message of the SageMaker HyperPod cluster.
InstanceGroups (list) --
The instance groups of the SageMaker HyperPod cluster.
(dict) --
Details of an instance group in a SageMaker HyperPod cluster.
CurrentCount (integer) --
The number of instances that are currently in the instance group of a SageMaker HyperPod cluster.
TargetCount (integer) --
The number of instances you specified to add to the instance group of a SageMaker HyperPod cluster.
MinCount (integer) --
The minimum number of instances that must be available in the instance group of a SageMaker HyperPod cluster before it transitions to InService status.
InstanceGroupName (string) --
The name of the instance group of a SageMaker HyperPod cluster.
InstanceType (string) --
The instance type of the instance group of a SageMaker HyperPod cluster.
LifeCycleConfig (dict) --
Details of LifeCycle configuration for the instance group.
SourceS3Uri (string) --
An Amazon S3 bucket path where your lifecycle scripts are stored.
OnCreate (string) --
The file name of the entrypoint script of lifecycle scripts under SourceS3Uri. This entrypoint script runs during cluster creation.
ExecutionRole (string) --
The execution role for the instance group to assume.
ThreadsPerCore (integer) --
The number you specified to TreadsPerCore in CreateCluster for enabling or disabling multithreading. For instance types that support multithreading, you can specify 1 for disabling multithreading and 2 for enabling multithreading. For more information, see the reference table of CPU cores and threads per CPU core per instance type in the Amazon Elastic Compute Cloud User Guide.
InstanceStorageConfigs (list) --
The additional storage configurations for the instances in the SageMaker HyperPod cluster instance group.
(dict) --
Defines the configuration for attaching additional storage to the instances in the SageMaker HyperPod cluster instance group. To learn more, see SageMaker HyperPod release notes: June 20, 2024.
EbsVolumeConfig (dict) --
Defines the configuration for attaching additional Amazon Elastic Block Store (EBS) volumes to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeSizeInGB (integer) --
The size in gigabytes (GB) of the additional EBS volume to be attached to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeKmsKeyId (string) --
The ID of a KMS key to encrypt the Amazon EBS volume.
RootVolume (boolean) --
Specifies whether the configuration is for the cluster's root or secondary Amazon EBS volume. You can specify two ClusterEbsVolumeConfig fields to configure both the root and secondary volumes. Set the value to True if you'd like to provide your own customer managed Amazon Web Services KMS key to encrypt the root volume. When True:
The configuration is applied to the root volume.
You can't specify the VolumeSizeInGB field. The size of the root volume is determined for you.
You must specify a KMS key ID for VolumeKmsKeyId to encrypt the root volume with your own KMS key instead of an Amazon Web Services owned KMS key.
Otherwise, by default, the value is False, and the following applies:
The configuration is applied to the secondary volume, while the root volume is encrypted with an Amazon Web Services owned key.
You must specify the VolumeSizeInGB field.
You can optionally specify the VolumeKmsKeyId to encrypt the secondary volume with your own KMS key instead of an Amazon Web Services owned KMS key.
OnStartDeepHealthChecks (list) --
A flag indicating whether deep health checks should be performed when the cluster instance group is created or updated.
(string) --
Status (string) --
The current status of the cluster instance group.
InService: The instance group is active and healthy.
Creating: The instance group is being provisioned.
Updating: The instance group is being updated.
Failed: The instance group has failed to provision or is no longer healthy.
Degraded: The instance group is degraded, meaning that some instances have failed to provision or are no longer healthy.
Deleting: The instance group is being deleted.
TrainingPlanArn (string) --
The Amazon Resource Name (ARN); of the training plan associated with this cluster instance group.
For more information about how to reserve GPU capacity for your SageMaker HyperPod clusters using Amazon SageMaker Training Plan, see ``CreateTrainingPlan ``.
TrainingPlanStatus (string) --
The current status of the training plan associated with this cluster instance group.
OverrideVpcConfig (dict) --
The customized Amazon VPC configuration at the instance group level that overrides the default Amazon VPC configuration of the SageMaker HyperPod cluster.
SecurityGroupIds (list) --
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) --
The ID of the subnets in the VPC to which you want to connect your training job or model. For information about the availability of specific instance types, see Supported Instance Types and Availability Zones.
(string) --
ScheduledUpdateConfig (dict) --
The configuration object of the schedule that SageMaker follows when updating the AMI.
ScheduleExpression (string) --
A cron expression that specifies the schedule that SageMaker follows when updating the AMI.
DeploymentConfig (dict) --
The configuration to use when updating the AMI versions.
RollingUpdatePolicy (dict) --
The policy that SageMaker uses when updating the AMI versions of the cluster.
MaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can update at a time.
Type (string) --
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) --
Specifies the amount or percentage of instances SageMaker updates at a time.
RollbackMaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can roll back at a time.
Type (string) --
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) --
Specifies the amount or percentage of instances SageMaker updates at a time.
WaitIntervalInSeconds (integer) --
The duration in seconds that SageMaker waits before updating more instances in the cluster.
AutoRollbackConfiguration (list) --
An array that contains the alarms that SageMaker monitors to know whether to roll back the AMI update.
(dict) --
The details of the alarm to monitor during the AMI update.
AlarmName (string) --
The name of the alarm.
CurrentImageId (string) --
The ID of the Amazon Machine Image (AMI) currently in use by the instance group.
DesiredImageId (string) --
The ID of the Amazon Machine Image (AMI) desired for the instance group.
ActiveOperations (dict) --
A map indicating active operations currently in progress for the instance group of a SageMaker HyperPod cluster. When there is a scaling operation in progress, this map contains a key Scaling with value 1.
(string) --
(integer) --
KubernetesConfig (dict) --
The Kubernetes configuration for the instance group that contains labels and taints to be applied for the nodes in this instance group.
CurrentLabels (dict) --
The current labels applied to cluster nodes of an instance group.
(string) --
(string) --
DesiredLabels (dict) --
The desired labels to be applied to cluster nodes of an instance group.
(string) --
(string) --
CurrentTaints (list) --
The current taints applied to cluster nodes of an instance group.
(dict) --
A Kubernetes taint that can be applied to cluster nodes.
Key (string) --
The key of the taint.
Value (string) --
The value of the taint.
Effect (string) --
The effect of the taint. Valid values are NoSchedule, PreferNoSchedule, and NoExecute.
DesiredTaints (list) --
The desired taints to be applied to cluster nodes of an instance group.
(dict) --
A Kubernetes taint that can be applied to cluster nodes.
Key (string) --
The key of the taint.
Value (string) --
The value of the taint.
Effect (string) --
The effect of the taint. Valid values are NoSchedule, PreferNoSchedule, and NoExecute.
CapacityRequirements (dict) --
The instance capacity requirements for the instance group.
Spot (dict) --
Configuration options specific to Spot instances.
OnDemand (dict) --
Configuration options specific to On-Demand instances.
TargetStateCount (integer) --
The number of nodes running a specific image ID since the last software update request.
SoftwareUpdateStatus (string) --
Status of the last software udpate request.
ActiveSoftwareUpdateConfig (dict) --
The configuration to use when updating the AMI versions.
RollingUpdatePolicy (dict) --
The policy that SageMaker uses when updating the AMI versions of the cluster.
MaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can update at a time.
Type (string) --
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) --
Specifies the amount or percentage of instances SageMaker updates at a time.
RollbackMaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can roll back at a time.
Type (string) --
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) --
Specifies the amount or percentage of instances SageMaker updates at a time.
WaitIntervalInSeconds (integer) --
The duration in seconds that SageMaker waits before updating more instances in the cluster.
AutoRollbackConfiguration (list) --
An array that contains the alarms that SageMaker monitors to know whether to roll back the AMI update.
(dict) --
The details of the alarm to monitor during the AMI update.
AlarmName (string) --
The name of the alarm.
RestrictedInstanceGroups (list) --
The specialized instance groups for training models like Amazon Nova to be created in the SageMaker HyperPod cluster.
(dict) --
The instance group details of the restricted instance group (RIG).
CurrentCount (integer) --
The number of instances that are currently in the restricted instance group of a SageMaker HyperPod cluster.
TargetCount (integer) --
The number of instances you specified to add to the restricted instance group of a SageMaker HyperPod cluster.
InstanceGroupName (string) --
The name of the restricted instance group of a SageMaker HyperPod cluster.
InstanceType (string) --
The instance type of the restricted instance group of a SageMaker HyperPod cluster.
ExecutionRole (string) --
The execution role for the restricted instance group to assume.
ThreadsPerCore (integer) --
The number you specified to TreadsPerCore in CreateCluster for enabling or disabling multithreading. For instance types that support multithreading, you can specify 1 for disabling multithreading and 2 for enabling multithreading. For more information, see the reference table of CPU cores and threads per CPU core per instance type in the Amazon Elastic Compute Cloud User Guide.
InstanceStorageConfigs (list) --
The additional storage configurations for the instances in the SageMaker HyperPod cluster restricted instance group.
(dict) --
Defines the configuration for attaching additional storage to the instances in the SageMaker HyperPod cluster instance group. To learn more, see SageMaker HyperPod release notes: June 20, 2024.
EbsVolumeConfig (dict) --
Defines the configuration for attaching additional Amazon Elastic Block Store (EBS) volumes to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeSizeInGB (integer) --
The size in gigabytes (GB) of the additional EBS volume to be attached to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeKmsKeyId (string) --
The ID of a KMS key to encrypt the Amazon EBS volume.
RootVolume (boolean) --
Specifies whether the configuration is for the cluster's root or secondary Amazon EBS volume. You can specify two ClusterEbsVolumeConfig fields to configure both the root and secondary volumes. Set the value to True if you'd like to provide your own customer managed Amazon Web Services KMS key to encrypt the root volume. When True:
The configuration is applied to the root volume.
You can't specify the VolumeSizeInGB field. The size of the root volume is determined for you.
You must specify a KMS key ID for VolumeKmsKeyId to encrypt the root volume with your own KMS key instead of an Amazon Web Services owned KMS key.
Otherwise, by default, the value is False, and the following applies:
The configuration is applied to the secondary volume, while the root volume is encrypted with an Amazon Web Services owned key.
You must specify the VolumeSizeInGB field.
You can optionally specify the VolumeKmsKeyId to encrypt the secondary volume with your own KMS key instead of an Amazon Web Services owned KMS key.
OnStartDeepHealthChecks (list) --
A flag indicating whether deep health checks should be performed when the cluster's restricted instance group is created or updated.
(string) --
Status (string) --
The current status of the cluster's restricted instance group.
InService: The restricted instance group is active and healthy.
Creating: The restricted instance group is being provisioned.
Updating: The restricted instance group is being updated.
Failed: The restricted instance group has failed to provision or is no longer healthy.
Degraded: The restricted instance group is degraded, meaning that some instances have failed to provision or are no longer healthy.
Deleting: The restricted instance group is being deleted.
TrainingPlanArn (string) --
The Amazon Resource Name (ARN) of the training plan to filter clusters by. For more information about reserving GPU capacity for your SageMaker HyperPod clusters using Amazon SageMaker Training Plan, see ``CreateTrainingPlan ``.
TrainingPlanStatus (string) --
The current status of the training plan associated with this cluster restricted instance group.
OverrideVpcConfig (dict) --
Specifies an Amazon Virtual Private Cloud (VPC) that your SageMaker jobs, hosted models, and compute resources have access to. You can control access to and from your resources by configuring a VPC. For more information, see Give SageMaker Access to Resources in your Amazon VPC.
SecurityGroupIds (list) --
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) --
The ID of the subnets in the VPC to which you want to connect your training job or model. For information about the availability of specific instance types, see Supported Instance Types and Availability Zones.
(string) --
ScheduledUpdateConfig (dict) --
The configuration object of the schedule that SageMaker follows when updating the AMI.
ScheduleExpression (string) --
A cron expression that specifies the schedule that SageMaker follows when updating the AMI.
DeploymentConfig (dict) --
The configuration to use when updating the AMI versions.
RollingUpdatePolicy (dict) --
The policy that SageMaker uses when updating the AMI versions of the cluster.
MaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can update at a time.
Type (string) --
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) --
Specifies the amount or percentage of instances SageMaker updates at a time.
RollbackMaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can roll back at a time.
Type (string) --
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) --
Specifies the amount or percentage of instances SageMaker updates at a time.
WaitIntervalInSeconds (integer) --
The duration in seconds that SageMaker waits before updating more instances in the cluster.
AutoRollbackConfiguration (list) --
An array that contains the alarms that SageMaker monitors to know whether to roll back the AMI update.
(dict) --
The details of the alarm to monitor during the AMI update.
AlarmName (string) --
The name of the alarm.
EnvironmentConfig (dict) --
The configuration for the restricted instance groups (RIG) environment.
FSxLustreConfig (dict) --
Configuration settings for an Amazon FSx for Lustre file system to be used with the cluster.
SizeInGiB (integer) --
The storage capacity of the Amazon FSx for Lustre file system, specified in gibibytes (GiB).
PerUnitStorageThroughput (integer) --
The throughput capacity of the Amazon FSx for Lustre file system, measured in MB/s per TiB of storage.
S3OutputPath (string) --
The Amazon S3 path where output data from the restricted instance group (RIG) environment will be stored.
VpcConfig (dict) --
Specifies an Amazon Virtual Private Cloud (VPC) that your SageMaker jobs, hosted models, and compute resources have access to. You can control access to and from your resources by configuring a VPC. For more information, see Give SageMaker Access to Resources in your Amazon VPC.
SecurityGroupIds (list) --
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) --
The ID of the subnets in the VPC to which you want to connect your training job or model. For information about the availability of specific instance types, see Supported Instance Types and Availability Zones.
(string) --
Orchestrator (dict) --
The type of orchestrator used for the SageMaker HyperPod cluster.
Eks (dict) --
The Amazon EKS cluster used as the orchestrator for the SageMaker HyperPod cluster.
ClusterArn (string) --
The Amazon Resource Name (ARN) of the Amazon EKS cluster associated with the SageMaker HyperPod cluster.
TieredStorageConfig (dict) --
The current configuration for managed tier checkpointing on the HyperPod cluster. For example, this shows whether the feature is enabled and the percentage of cluster memory allocated for checkpoint storage.
Mode (string) --
Specifies whether managed tier checkpointing is enabled or disabled for the HyperPod cluster. When set to Enable, the system installs a memory management daemon that provides disaggregated memory as a service for checkpoint storage. When set to Disable, the feature is turned off and the memory management daemon is removed from the cluster.
InstanceMemoryAllocationPercentage (integer) --
The percentage (int) of cluster memory to allocate for checkpointing.
NodeRecovery (string) --
The node recovery mode configured for the SageMaker HyperPod cluster.
NodeProvisioningMode (string) --
The mode used for provisioning nodes in the cluster.
ClusterRole (string) --
The Amazon Resource Name (ARN) of the IAM role that HyperPod uses for cluster autoscaling operations.
AutoScaling (dict) --
The current autoscaling configuration and status for the autoscaler.
Mode (string) --
Describes whether autoscaling is enabled or disabled for the cluster.
AutoScalerType (string) --
The type of autoscaler configured for the cluster.
Status (string) --
The current status of the autoscaling configuration. Valid values are InService, Failed, Creating, and Deleting.
FailureMessage (string) --
If the autoscaling status is Failed, this field contains a message describing the failure.
{'EventDetails': {'EventDetails': {'EventMetadata': {'InstanceGroupScaling': {'MinCount': 'integer'}}}}}
Retrieves detailed information about a specific event for a given HyperPod cluster. This functionality is only supported when the NodeProvisioningMode is set to Continuous.
See also: AWS API Documentation
Request Syntax
client.describe_cluster_event(
EventId='string',
ClusterName='string'
)
string
[REQUIRED]
The unique identifier (UUID) of the event to describe. This ID can be obtained from the ListClusterEvents operation.
string
[REQUIRED]
The name or Amazon Resource Name (ARN) of the HyperPod cluster associated with the event.
dict
Response Syntax
{
'EventDetails': {
'EventId': 'string',
'ClusterArn': 'string',
'ClusterName': 'string',
'InstanceGroupName': 'string',
'InstanceId': 'string',
'ResourceType': 'Cluster'|'InstanceGroup'|'Instance',
'EventTime': datetime(2015, 1, 1),
'EventDetails': {
'EventMetadata': {
'Cluster': {
'FailureMessage': 'string',
'EksRoleAccessEntries': [
'string',
],
'SlrAccessEntry': 'string'
},
'InstanceGroup': {
'FailureMessage': 'string',
'AvailabilityZoneId': 'string',
'CapacityReservation': {
'Arn': 'string',
'Type': 'ODCR'|'CRG'
},
'SubnetId': 'string',
'SecurityGroupIds': [
'string',
],
'AmiOverride': 'string'
},
'InstanceGroupScaling': {
'InstanceCount': 123,
'TargetCount': 123,
'MinCount': 123,
'FailureMessage': 'string'
},
'Instance': {
'CustomerEni': 'string',
'AdditionalEnis': {
'EfaEnis': [
'string',
]
},
'CapacityReservation': {
'Arn': 'string',
'Type': 'ODCR'|'CRG'
},
'FailureMessage': 'string',
'LcsExecutionState': 'string',
'NodeLogicalId': 'string'
}
}
},
'Description': 'string'
}
}
Response Structure
(dict) --
EventDetails (dict) --
Detailed information about the requested cluster event, including event metadata for various resource types such as Cluster, InstanceGroup, Instance, and their associated attributes.
EventId (string) --
The unique identifier (UUID) of the event.
ClusterArn (string) --
The Amazon Resource Name (ARN) of the HyperPod cluster associated with the event.
ClusterName (string) --
The name of the HyperPod cluster associated with the event.
InstanceGroupName (string) --
The name of the instance group associated with the event, if applicable.
InstanceId (string) --
The EC2 instance ID associated with the event, if applicable.
ResourceType (string) --
The type of resource associated with the event. Valid values are Cluster, InstanceGroup, or Instance.
EventTime (datetime) --
The timestamp when the event occurred.
EventDetails (dict) --
Additional details about the event, including event-specific metadata.
EventMetadata (dict) --
Metadata specific to the event, which may include information about the cluster, instance group, or instance involved.
Cluster (dict) --
Metadata specific to cluster-level events.
FailureMessage (string) --
An error message describing why the cluster level operation (such as creating, updating, or deleting) failed.
EksRoleAccessEntries (list) --
A list of Amazon EKS IAM role ARNs associated with the cluster. This is created by HyperPod on your behalf and only applies for EKS orchestrated clusters.
(string) --
SlrAccessEntry (string) --
The Service-Linked Role (SLR) associated with the cluster. This is created by HyperPod on your behalf and only applies for EKS orchestrated clusters.
InstanceGroup (dict) --
Metadata specific to instance group-level events.
FailureMessage (string) --
An error message describing why the instance group level operation (such as creating, scaling, or deleting) failed.
AvailabilityZoneId (string) --
The ID of the Availability Zone where the instance group is located.
CapacityReservation (dict) --
Information about the Capacity Reservation used by the instance group.
Arn (string) --
The Amazon Resource Name (ARN) of the Capacity Reservation.
Type (string) --
The type of Capacity Reservation. Valid values are ODCR (On-Demand Capacity Reservation) or CRG (Capacity Reservation Group).
SubnetId (string) --
The ID of the subnet where the instance group is located.
SecurityGroupIds (list) --
A list of security group IDs associated with the instance group.
(string) --
AmiOverride (string) --
If you use a custom Amazon Machine Image (AMI) for the instance group, this field shows the ID of the custom AMI.
InstanceGroupScaling (dict) --
Metadata related to instance group scaling events.
InstanceCount (integer) --
The current number of instances in the group.
TargetCount (integer) --
The desired number of instances for the group after scaling.
MinCount (integer) --
Minimum instance count of the instance group.
FailureMessage (string) --
An error message describing why the scaling operation failed, if applicable.
Instance (dict) --
Metadata specific to instance-level events.
CustomerEni (string) --
The ID of the customer-managed Elastic Network Interface (ENI) associated with the instance.
AdditionalEnis (dict) --
Information about additional Elastic Network Interfaces (ENIs) associated with the instance.
EfaEnis (list) --
A list of Elastic Fabric Adapter (EFA) ENIs associated with the instance.
(string) --
CapacityReservation (dict) --
Information about the Capacity Reservation used by the instance.
Arn (string) --
The Amazon Resource Name (ARN) of the Capacity Reservation.
Type (string) --
The type of Capacity Reservation. Valid values are ODCR (On-Demand Capacity Reservation) or CRG (Capacity Reservation Group).
FailureMessage (string) --
An error message describing why the instance creation or update failed, if applicable.
LcsExecutionState (string) --
The execution state of the Lifecycle Script (LCS) for the instance.
NodeLogicalId (string) --
The unique logical identifier of the node within the cluster. The ID used here is the same object as in the BatchAddClusterNodes API.
Description (string) --
A human-readable description of the event.
{'NodeDetails': {'CapacityType': 'Spot | OnDemand',
'KubernetesConfig': {'CurrentLabels': {'string': 'string'},
'CurrentTaints': [{'Effect': 'NoSchedule '
'| '
'PreferNoSchedule '
'| '
'NoExecute',
'Key': 'string',
'Value': 'string'}],
'DesiredLabels': {'string': 'string'},
'DesiredTaints': [{'Effect': 'NoSchedule '
'| '
'PreferNoSchedule '
'| '
'NoExecute',
'Key': 'string',
'Value': 'string'}]}}}
Retrieves information of a node (also called a instance interchangeably) of a SageMaker HyperPod cluster.
See also: AWS API Documentation
Request Syntax
client.describe_cluster_node(
ClusterName='string',
NodeId='string',
NodeLogicalId='string'
)
string
[REQUIRED]
The string name or the Amazon Resource Name (ARN) of the SageMaker HyperPod cluster in which the node is.
string
The ID of the SageMaker HyperPod cluster node.
string
The logical identifier of the node to describe. You can specify either NodeLogicalId or InstanceId, but not both. NodeLogicalId can be used to describe nodes that are still being provisioned and don't yet have an InstanceId assigned.
dict
Response Syntax
{
'NodeDetails': {
'InstanceGroupName': 'string',
'InstanceId': 'string',
'NodeLogicalId': 'string',
'InstanceStatus': {
'Status': 'Running'|'Failure'|'Pending'|'ShuttingDown'|'SystemUpdating'|'DeepHealthCheckInProgress'|'NotFound',
'Message': 'string'
},
'InstanceType': 'ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p6e-gb200.36xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.c5.large'|'ml.c5.xlarge'|'ml.c5.2xlarge'|'ml.c5.4xlarge'|'ml.c5.9xlarge'|'ml.c5.12xlarge'|'ml.c5.18xlarge'|'ml.c5.24xlarge'|'ml.c5n.large'|'ml.c5n.2xlarge'|'ml.c5n.4xlarge'|'ml.c5n.9xlarge'|'ml.c5n.18xlarge'|'ml.m5.large'|'ml.m5.xlarge'|'ml.m5.2xlarge'|'ml.m5.4xlarge'|'ml.m5.8xlarge'|'ml.m5.12xlarge'|'ml.m5.16xlarge'|'ml.m5.24xlarge'|'ml.t3.medium'|'ml.t3.large'|'ml.t3.xlarge'|'ml.t3.2xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.16xlarge'|'ml.g6.12xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.gr6.4xlarge'|'ml.gr6.8xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.16xlarge'|'ml.g6e.12xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.p6-b200.48xlarge'|'ml.trn2.3xlarge'|'ml.trn2.48xlarge'|'ml.c6i.large'|'ml.c6i.xlarge'|'ml.c6i.2xlarge'|'ml.c6i.4xlarge'|'ml.c6i.8xlarge'|'ml.c6i.12xlarge'|'ml.c6i.16xlarge'|'ml.c6i.24xlarge'|'ml.c6i.32xlarge'|'ml.m6i.large'|'ml.m6i.xlarge'|'ml.m6i.2xlarge'|'ml.m6i.4xlarge'|'ml.m6i.8xlarge'|'ml.m6i.12xlarge'|'ml.m6i.16xlarge'|'ml.m6i.24xlarge'|'ml.m6i.32xlarge'|'ml.r6i.large'|'ml.r6i.xlarge'|'ml.r6i.2xlarge'|'ml.r6i.4xlarge'|'ml.r6i.8xlarge'|'ml.r6i.12xlarge'|'ml.r6i.16xlarge'|'ml.r6i.24xlarge'|'ml.r6i.32xlarge'|'ml.i3en.large'|'ml.i3en.xlarge'|'ml.i3en.2xlarge'|'ml.i3en.3xlarge'|'ml.i3en.6xlarge'|'ml.i3en.12xlarge'|'ml.i3en.24xlarge'|'ml.m7i.large'|'ml.m7i.xlarge'|'ml.m7i.2xlarge'|'ml.m7i.4xlarge'|'ml.m7i.8xlarge'|'ml.m7i.12xlarge'|'ml.m7i.16xlarge'|'ml.m7i.24xlarge'|'ml.m7i.48xlarge'|'ml.r7i.large'|'ml.r7i.xlarge'|'ml.r7i.2xlarge'|'ml.r7i.4xlarge'|'ml.r7i.8xlarge'|'ml.r7i.12xlarge'|'ml.r7i.16xlarge'|'ml.r7i.24xlarge'|'ml.r7i.48xlarge',
'LaunchTime': datetime(2015, 1, 1),
'LastSoftwareUpdateTime': datetime(2015, 1, 1),
'LifeCycleConfig': {
'SourceS3Uri': 'string',
'OnCreate': 'string'
},
'OverrideVpcConfig': {
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
},
'ThreadsPerCore': 123,
'InstanceStorageConfigs': [
{
'EbsVolumeConfig': {
'VolumeSizeInGB': 123,
'VolumeKmsKeyId': 'string',
'RootVolume': True|False
}
},
],
'PrivatePrimaryIp': 'string',
'PrivatePrimaryIpv6': 'string',
'PrivateDnsHostname': 'string',
'Placement': {
'AvailabilityZone': 'string',
'AvailabilityZoneId': 'string'
},
'CurrentImageId': 'string',
'DesiredImageId': 'string',
'UltraServerInfo': {
'Id': 'string'
},
'KubernetesConfig': {
'CurrentLabels': {
'string': 'string'
},
'DesiredLabels': {
'string': 'string'
},
'CurrentTaints': [
{
'Key': 'string',
'Value': 'string',
'Effect': 'NoSchedule'|'PreferNoSchedule'|'NoExecute'
},
],
'DesiredTaints': [
{
'Key': 'string',
'Value': 'string',
'Effect': 'NoSchedule'|'PreferNoSchedule'|'NoExecute'
},
]
},
'CapacityType': 'Spot'|'OnDemand'
}
}
Response Structure
(dict) --
NodeDetails (dict) --
The details of the SageMaker HyperPod cluster node.
InstanceGroupName (string) --
The instance group name in which the instance is.
InstanceId (string) --
The ID of the instance.
NodeLogicalId (string) --
A unique identifier for the node that persists throughout its lifecycle, from provisioning request to termination. This identifier can be used to track the node even before it has an assigned InstanceId.
InstanceStatus (dict) --
The status of the instance.
Status (string) --
The status of an instance in a SageMaker HyperPod cluster.
Message (string) --
The message from an instance in a SageMaker HyperPod cluster.
InstanceType (string) --
The type of the instance.
LaunchTime (datetime) --
The time when the instance is launched.
LastSoftwareUpdateTime (datetime) --
The time when the cluster was last updated.
LifeCycleConfig (dict) --
The LifeCycle configuration applied to the instance.
SourceS3Uri (string) --
An Amazon S3 bucket path where your lifecycle scripts are stored.
OnCreate (string) --
The file name of the entrypoint script of lifecycle scripts under SourceS3Uri. This entrypoint script runs during cluster creation.
OverrideVpcConfig (dict) --
The customized Amazon VPC configuration at the instance group level that overrides the default Amazon VPC configuration of the SageMaker HyperPod cluster.
SecurityGroupIds (list) --
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) --
The ID of the subnets in the VPC to which you want to connect your training job or model. For information about the availability of specific instance types, see Supported Instance Types and Availability Zones.
(string) --
ThreadsPerCore (integer) --
The number of threads per CPU core you specified under CreateCluster.
InstanceStorageConfigs (list) --
The configurations of additional storage specified to the instance group where the instance (node) is launched.
(dict) --
Defines the configuration for attaching additional storage to the instances in the SageMaker HyperPod cluster instance group. To learn more, see SageMaker HyperPod release notes: June 20, 2024.
EbsVolumeConfig (dict) --
Defines the configuration for attaching additional Amazon Elastic Block Store (EBS) volumes to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeSizeInGB (integer) --
The size in gigabytes (GB) of the additional EBS volume to be attached to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeKmsKeyId (string) --
The ID of a KMS key to encrypt the Amazon EBS volume.
RootVolume (boolean) --
Specifies whether the configuration is for the cluster's root or secondary Amazon EBS volume. You can specify two ClusterEbsVolumeConfig fields to configure both the root and secondary volumes. Set the value to True if you'd like to provide your own customer managed Amazon Web Services KMS key to encrypt the root volume. When True:
The configuration is applied to the root volume.
You can't specify the VolumeSizeInGB field. The size of the root volume is determined for you.
You must specify a KMS key ID for VolumeKmsKeyId to encrypt the root volume with your own KMS key instead of an Amazon Web Services owned KMS key.
Otherwise, by default, the value is False, and the following applies:
The configuration is applied to the secondary volume, while the root volume is encrypted with an Amazon Web Services owned key.
You must specify the VolumeSizeInGB field.
You can optionally specify the VolumeKmsKeyId to encrypt the secondary volume with your own KMS key instead of an Amazon Web Services owned KMS key.
PrivatePrimaryIp (string) --
The private primary IP address of the SageMaker HyperPod cluster node.
PrivatePrimaryIpv6 (string) --
The private primary IPv6 address of the SageMaker HyperPod cluster node when configured with an Amazon VPC that supports IPv6 and includes subnets with IPv6 addressing enabled in either the cluster Amazon VPC configuration or the instance group Amazon VPC configuration.
PrivateDnsHostname (string) --
The private DNS hostname of the SageMaker HyperPod cluster node.
Placement (dict) --
The placement details of the SageMaker HyperPod cluster node.
AvailabilityZone (string) --
The Availability Zone where the node in the SageMaker HyperPod cluster is launched.
AvailabilityZoneId (string) --
The unique identifier (ID) of the Availability Zone where the node in the SageMaker HyperPod cluster is launched.
CurrentImageId (string) --
The ID of the Amazon Machine Image (AMI) currently in use by the node.
DesiredImageId (string) --
The ID of the Amazon Machine Image (AMI) desired for the node.
UltraServerInfo (dict) --
Contains information about the UltraServer.
Id (string) --
The unique identifier of the UltraServer.
KubernetesConfig (dict) --
The Kubernetes configuration applied to this node, showing both the current and desired state of labels and taints. The cluster works to reconcile the actual state with the declared state.
CurrentLabels (dict) --
The current labels applied to the cluster node.
(string) --
(string) --
DesiredLabels (dict) --
The desired labels to be applied to the cluster node.
(string) --
(string) --
CurrentTaints (list) --
The current taints applied to the cluster node.
(dict) --
A Kubernetes taint that can be applied to cluster nodes.
Key (string) --
The key of the taint.
Value (string) --
The value of the taint.
Effect (string) --
The effect of the taint. Valid values are NoSchedule, PreferNoSchedule, and NoExecute.
DesiredTaints (list) --
The desired taints to be applied to the cluster node.
(dict) --
A Kubernetes taint that can be applied to cluster nodes.
Key (string) --
The key of the taint.
Value (string) --
The value of the taint.
Effect (string) --
The effect of the taint. Valid values are NoSchedule, PreferNoSchedule, and NoExecute.
CapacityType (string) --
The capacity type of the node. Valid values are OnDemand and Spot. When set to OnDemand, the node is launched as an On-Demand instance. When set to Spot, the node is launched as a Spot instance.
{'DeploymentInstanceType': {'ml.p5en.48xlarge', 'ml.p5e.48xlarge'},
'MaxInstanceCount': 'integer',
'ModelSource': {'SageMakerModel': {'ModelName': 'string'}},
'OptimizationConfigs': {'ModelSpeculativeDecodingConfig': {'Technique': 'EAGLE',
'TrainingDataSource': {'S3DataType': 'S3Prefix '
'| '
'ManifestFile',
'S3Uri': 'string'}}},
'OutputConfig': {'SageMakerModel': {'ModelName': 'string'}}}
Provides the properties of the specified optimization job.
See also: AWS API Documentation
Request Syntax
client.describe_optimization_job(
OptimizationJobName='string'
)
string
[REQUIRED]
The name that you assigned to the optimization job.
dict
Response Syntax
{
'OptimizationJobArn': 'string',
'OptimizationJobStatus': 'INPROGRESS'|'COMPLETED'|'FAILED'|'STARTING'|'STOPPING'|'STOPPED',
'OptimizationStartTime': datetime(2015, 1, 1),
'OptimizationEndTime': datetime(2015, 1, 1),
'CreationTime': datetime(2015, 1, 1),
'LastModifiedTime': datetime(2015, 1, 1),
'FailureReason': 'string',
'OptimizationJobName': 'string',
'ModelSource': {
'S3': {
'S3Uri': 'string',
'ModelAccessConfig': {
'AcceptEula': True|False
}
},
'SageMakerModel': {
'ModelName': 'string'
}
},
'OptimizationEnvironment': {
'string': 'string'
},
'DeploymentInstanceType': 'ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.12xlarge'|'ml.g6.16xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.12xlarge'|'ml.g6e.16xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.inf2.xlarge'|'ml.inf2.8xlarge'|'ml.inf2.24xlarge'|'ml.inf2.48xlarge'|'ml.trn1.2xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge',
'MaxInstanceCount': 123,
'OptimizationConfigs': [
{
'ModelQuantizationConfig': {
'Image': 'string',
'OverrideEnvironment': {
'string': 'string'
}
},
'ModelCompilationConfig': {
'Image': 'string',
'OverrideEnvironment': {
'string': 'string'
}
},
'ModelShardingConfig': {
'Image': 'string',
'OverrideEnvironment': {
'string': 'string'
}
},
'ModelSpeculativeDecodingConfig': {
'Technique': 'EAGLE',
'TrainingDataSource': {
'S3Uri': 'string',
'S3DataType': 'S3Prefix'|'ManifestFile'
}
}
},
],
'OutputConfig': {
'KmsKeyId': 'string',
'S3OutputLocation': 'string',
'SageMakerModel': {
'ModelName': 'string'
}
},
'OptimizationOutput': {
'RecommendedInferenceImage': 'string'
},
'RoleArn': 'string',
'StoppingCondition': {
'MaxRuntimeInSeconds': 123,
'MaxWaitTimeInSeconds': 123,
'MaxPendingTimeInSeconds': 123
},
'VpcConfig': {
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
}
}
Response Structure
(dict) --
OptimizationJobArn (string) --
The Amazon Resource Name (ARN) of the optimization job.
OptimizationJobStatus (string) --
The current status of the optimization job.
OptimizationStartTime (datetime) --
The time when the optimization job started.
OptimizationEndTime (datetime) --
The time when the optimization job finished processing.
CreationTime (datetime) --
The time when you created the optimization job.
LastModifiedTime (datetime) --
The time when the optimization job was last updated.
FailureReason (string) --
If the optimization job status is FAILED, the reason for the failure.
OptimizationJobName (string) --
The name that you assigned to the optimization job.
ModelSource (dict) --
The location of the source model to optimize with an optimization job.
S3 (dict) --
The Amazon S3 location of a source model to optimize with an optimization job.
S3Uri (string) --
An Amazon S3 URI that locates a source model to optimize with an optimization job.
ModelAccessConfig (dict) --
The access configuration settings for the source ML model for an optimization job, where you can accept the model end-user license agreement (EULA).
AcceptEula (boolean) --
Specifies agreement to the model end-user license agreement (EULA). The AcceptEula value must be explicitly defined as True in order to accept the EULA that this model requires. You are responsible for reviewing and complying with any applicable license terms and making sure they are acceptable for your use case before downloading or using a model.
SageMakerModel (dict) --
The name of an existing SageMaker model to optimize with an optimization job.
ModelName (string) --
The name of a SageMaker model.
OptimizationEnvironment (dict) --
The environment variables to set in the model container.
(string) --
(string) --
DeploymentInstanceType (string) --
The type of instance that hosts the optimized model that you create with the optimization job.
MaxInstanceCount (integer) --
The maximum number of instances to use for the optimization job.
OptimizationConfigs (list) --
Settings for each of the optimization techniques that the job applies.
(dict) --
Settings for an optimization technique that you apply with a model optimization job.
ModelQuantizationConfig (dict) --
Settings for the model quantization technique that's applied by a model optimization job.
Image (string) --
The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.
OverrideEnvironment (dict) --
Environment variables that override the default ones in the model container.
(string) --
(string) --
ModelCompilationConfig (dict) --
Settings for the model compilation technique that's applied by a model optimization job.
Image (string) --
The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.
OverrideEnvironment (dict) --
Environment variables that override the default ones in the model container.
(string) --
(string) --
ModelShardingConfig (dict) --
Settings for the model sharding technique that's applied by a model optimization job.
Image (string) --
The URI of an LMI DLC in Amazon ECR. SageMaker uses this image to run the optimization.
OverrideEnvironment (dict) --
Environment variables that override the default ones in the model container.
(string) --
(string) --
ModelSpeculativeDecodingConfig (dict) --
Settings for the model speculative decoding technique that's applied by a model optimization job.
Technique (string) --
The speculative decoding technique to apply during model optimization.
TrainingDataSource (dict) --
The location of the training data to use for speculative decoding. The data must be formatted as ShareGPT, OpenAI Completions or OpenAI Chat Completions. The input can also be unencrypted captured data from a SageMaker endpoint as long as the endpoint uses one of the above formats.
S3Uri (string) --
The Amazon S3 URI that points to the training data for speculative decoding.
S3DataType (string) --
The type of data stored in the Amazon S3 location. Valid values are S3Prefix or ManifestFile.
OutputConfig (dict) --
Details for where to store the optimized model that you create with the optimization job.
KmsKeyId (string) --
The Amazon Resource Name (ARN) of a key in Amazon Web Services KMS. SageMaker uses they key to encrypt the artifacts of the optimized model when SageMaker uploads the model to Amazon S3.
S3OutputLocation (string) --
The Amazon S3 URI for where to store the optimized model that you create with an optimization job.
SageMakerModel (dict) --
The name of a SageMaker model to use as the output destination for an optimization job.
ModelName (string) --
The name of a SageMaker model.
OptimizationOutput (dict) --
Output values produced by an optimization job.
RecommendedInferenceImage (string) --
The image that SageMaker recommends that you use to host the optimized model that you created with an optimization job.
RoleArn (string) --
The ARN of the IAM role that you assigned to the optimization job.
StoppingCondition (dict) --
Specifies a limit to how long a job can run. When the job reaches the time limit, SageMaker ends the job. Use this API to cap costs.
To stop a training job, SageMaker sends the algorithm the SIGTERM signal, which delays job termination for 120 seconds. Algorithms can use this 120-second window to save the model artifacts, so the results of training are not lost.
The training algorithms provided by SageMaker automatically save the intermediate results of a model training job when possible. This attempt to save artifacts is only a best effort case as model might not be in a state from which it can be saved. For example, if training has just started, the model might not be ready to save. When saved, this intermediate data is a valid model artifact. You can use it to create a model with CreateModel.
MaxRuntimeInSeconds (integer) --
The maximum length of time, in seconds, that a training or compilation job can run before it is stopped.
For compilation jobs, if the job does not complete during this time, a TimeOut error is generated. We recommend starting with 900 seconds and increasing as necessary based on your model.
For all other jobs, if the job does not complete during this time, SageMaker ends the job. When RetryStrategy is specified in the job request, MaxRuntimeInSeconds specifies the maximum time for all of the attempts in total, not each individual attempt. The default value is 1 day. The maximum value is 28 days.
The maximum time that a TrainingJob can run in total, including any time spent publishing metrics or archiving and uploading models after it has been stopped, is 30 days.
MaxWaitTimeInSeconds (integer) --
The maximum length of time, in seconds, that a managed Spot training job has to complete. It is the amount of time spent waiting for Spot capacity plus the amount of time the job can run. It must be equal to or greater than MaxRuntimeInSeconds. If the job does not complete during this time, SageMaker ends the job.
When RetryStrategy is specified in the job request, MaxWaitTimeInSeconds specifies the maximum time for all of the attempts in total, not each individual attempt.
MaxPendingTimeInSeconds (integer) --
The maximum length of time, in seconds, that a training or compilation job can be pending before it is stopped.
VpcConfig (dict) --
A VPC in Amazon VPC that your optimized model has access to.
SecurityGroupIds (list) --
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) --
The ID of the subnets in the VPC to which you want to connect your optimized model.
(string) --
{'OptimizationJobSummaries': {'DeploymentInstanceType': {'ml.p5e.48xlarge',
'ml.p5en.48xlarge'},
'MaxInstanceCount': 'integer'}}
Lists the optimization jobs in your account and their properties.
See also: AWS API Documentation
Request Syntax
client.list_optimization_jobs(
NextToken='string',
MaxResults=123,
CreationTimeAfter=datetime(2015, 1, 1),
CreationTimeBefore=datetime(2015, 1, 1),
LastModifiedTimeAfter=datetime(2015, 1, 1),
LastModifiedTimeBefore=datetime(2015, 1, 1),
OptimizationContains='string',
NameContains='string',
StatusEquals='INPROGRESS'|'COMPLETED'|'FAILED'|'STARTING'|'STOPPING'|'STOPPED',
SortBy='Name'|'CreationTime'|'Status',
SortOrder='Ascending'|'Descending'
)
string
A token that you use to get the next set of results following a truncated response. If the response to the previous request was truncated, that response provides the value for this token.
integer
The maximum number of optimization jobs to return in the response. The default is 50.
datetime
Filters the results to only those optimization jobs that were created after the specified time.
datetime
Filters the results to only those optimization jobs that were created before the specified time.
datetime
Filters the results to only those optimization jobs that were updated after the specified time.
datetime
Filters the results to only those optimization jobs that were updated before the specified time.
string
Filters the results to only those optimization jobs that apply the specified optimization techniques. You can specify either Quantization or Compilation.
string
Filters the results to only those optimization jobs with a name that contains the specified string.
string
Filters the results to only those optimization jobs with the specified status.
string
The field by which to sort the optimization jobs in the response. The default is CreationTime
string
The sort order for results. The default is Ascending
dict
Response Syntax
{
'OptimizationJobSummaries': [
{
'OptimizationJobName': 'string',
'OptimizationJobArn': 'string',
'CreationTime': datetime(2015, 1, 1),
'OptimizationJobStatus': 'INPROGRESS'|'COMPLETED'|'FAILED'|'STARTING'|'STOPPING'|'STOPPED',
'OptimizationStartTime': datetime(2015, 1, 1),
'OptimizationEndTime': datetime(2015, 1, 1),
'LastModifiedTime': datetime(2015, 1, 1),
'DeploymentInstanceType': 'ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.12xlarge'|'ml.g6.16xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.12xlarge'|'ml.g6e.16xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.inf2.xlarge'|'ml.inf2.8xlarge'|'ml.inf2.24xlarge'|'ml.inf2.48xlarge'|'ml.trn1.2xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge',
'MaxInstanceCount': 123,
'OptimizationTypes': [
'string',
]
},
],
'NextToken': 'string'
}
Response Structure
(dict) --
OptimizationJobSummaries (list) --
A list of optimization jobs and their properties that matches any of the filters you specified in the request.
(dict) --
Summarizes an optimization job by providing some of its key properties.
OptimizationJobName (string) --
The name that you assigned to the optimization job.
OptimizationJobArn (string) --
The Amazon Resource Name (ARN) of the optimization job.
CreationTime (datetime) --
The time when you created the optimization job.
OptimizationJobStatus (string) --
The current status of the optimization job.
OptimizationStartTime (datetime) --
The time when the optimization job started.
OptimizationEndTime (datetime) --
The time when the optimization job finished processing.
LastModifiedTime (datetime) --
The time when the optimization job was last updated.
DeploymentInstanceType (string) --
The type of instance that hosts the optimized model that you create with the optimization job.
MaxInstanceCount (integer) --
The maximum number of instances to use for the optimization job.
OptimizationTypes (list) --
The optimization techniques that are applied by the optimization job.
(string) --
NextToken (string) --
The token to use in a subsequent request to get the next set of results following a truncated response.
{'InstanceGroups': {'CapacityRequirements': {'OnDemand': {}, 'Spot': {}},
'KubernetesConfig': {'Labels': {'string': 'string'},
'Taints': [{'Effect': 'NoSchedule | '
'PreferNoSchedule '
'| NoExecute',
'Key': 'string',
'Value': 'string'}]},
'MinInstanceCount': 'integer'}}
Updates a SageMaker HyperPod cluster.
See also: AWS API Documentation
Request Syntax
client.update_cluster(
ClusterName='string',
InstanceGroups=[
{
'InstanceCount': 123,
'MinInstanceCount': 123,
'InstanceGroupName': 'string',
'InstanceType': 'ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p6e-gb200.36xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.c5.large'|'ml.c5.xlarge'|'ml.c5.2xlarge'|'ml.c5.4xlarge'|'ml.c5.9xlarge'|'ml.c5.12xlarge'|'ml.c5.18xlarge'|'ml.c5.24xlarge'|'ml.c5n.large'|'ml.c5n.2xlarge'|'ml.c5n.4xlarge'|'ml.c5n.9xlarge'|'ml.c5n.18xlarge'|'ml.m5.large'|'ml.m5.xlarge'|'ml.m5.2xlarge'|'ml.m5.4xlarge'|'ml.m5.8xlarge'|'ml.m5.12xlarge'|'ml.m5.16xlarge'|'ml.m5.24xlarge'|'ml.t3.medium'|'ml.t3.large'|'ml.t3.xlarge'|'ml.t3.2xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.16xlarge'|'ml.g6.12xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.gr6.4xlarge'|'ml.gr6.8xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.16xlarge'|'ml.g6e.12xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.p6-b200.48xlarge'|'ml.trn2.3xlarge'|'ml.trn2.48xlarge'|'ml.c6i.large'|'ml.c6i.xlarge'|'ml.c6i.2xlarge'|'ml.c6i.4xlarge'|'ml.c6i.8xlarge'|'ml.c6i.12xlarge'|'ml.c6i.16xlarge'|'ml.c6i.24xlarge'|'ml.c6i.32xlarge'|'ml.m6i.large'|'ml.m6i.xlarge'|'ml.m6i.2xlarge'|'ml.m6i.4xlarge'|'ml.m6i.8xlarge'|'ml.m6i.12xlarge'|'ml.m6i.16xlarge'|'ml.m6i.24xlarge'|'ml.m6i.32xlarge'|'ml.r6i.large'|'ml.r6i.xlarge'|'ml.r6i.2xlarge'|'ml.r6i.4xlarge'|'ml.r6i.8xlarge'|'ml.r6i.12xlarge'|'ml.r6i.16xlarge'|'ml.r6i.24xlarge'|'ml.r6i.32xlarge'|'ml.i3en.large'|'ml.i3en.xlarge'|'ml.i3en.2xlarge'|'ml.i3en.3xlarge'|'ml.i3en.6xlarge'|'ml.i3en.12xlarge'|'ml.i3en.24xlarge'|'ml.m7i.large'|'ml.m7i.xlarge'|'ml.m7i.2xlarge'|'ml.m7i.4xlarge'|'ml.m7i.8xlarge'|'ml.m7i.12xlarge'|'ml.m7i.16xlarge'|'ml.m7i.24xlarge'|'ml.m7i.48xlarge'|'ml.r7i.large'|'ml.r7i.xlarge'|'ml.r7i.2xlarge'|'ml.r7i.4xlarge'|'ml.r7i.8xlarge'|'ml.r7i.12xlarge'|'ml.r7i.16xlarge'|'ml.r7i.24xlarge'|'ml.r7i.48xlarge',
'LifeCycleConfig': {
'SourceS3Uri': 'string',
'OnCreate': 'string'
},
'ExecutionRole': 'string',
'ThreadsPerCore': 123,
'InstanceStorageConfigs': [
{
'EbsVolumeConfig': {
'VolumeSizeInGB': 123,
'VolumeKmsKeyId': 'string',
'RootVolume': True|False
}
},
],
'OnStartDeepHealthChecks': [
'InstanceStress'|'InstanceConnectivity',
],
'TrainingPlanArn': 'string',
'OverrideVpcConfig': {
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
},
'ScheduledUpdateConfig': {
'ScheduleExpression': 'string',
'DeploymentConfig': {
'RollingUpdatePolicy': {
'MaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
},
'RollbackMaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
}
},
'WaitIntervalInSeconds': 123,
'AutoRollbackConfiguration': [
{
'AlarmName': 'string'
},
]
}
},
'ImageId': 'string',
'KubernetesConfig': {
'Labels': {
'string': 'string'
},
'Taints': [
{
'Key': 'string',
'Value': 'string',
'Effect': 'NoSchedule'|'PreferNoSchedule'|'NoExecute'
},
]
},
'CapacityRequirements': {
'Spot': {}
,
'OnDemand': {}
}
},
],
RestrictedInstanceGroups=[
{
'InstanceCount': 123,
'InstanceGroupName': 'string',
'InstanceType': 'ml.p4d.24xlarge'|'ml.p4de.24xlarge'|'ml.p5.48xlarge'|'ml.p6e-gb200.36xlarge'|'ml.trn1.32xlarge'|'ml.trn1n.32xlarge'|'ml.g5.xlarge'|'ml.g5.2xlarge'|'ml.g5.4xlarge'|'ml.g5.8xlarge'|'ml.g5.12xlarge'|'ml.g5.16xlarge'|'ml.g5.24xlarge'|'ml.g5.48xlarge'|'ml.c5.large'|'ml.c5.xlarge'|'ml.c5.2xlarge'|'ml.c5.4xlarge'|'ml.c5.9xlarge'|'ml.c5.12xlarge'|'ml.c5.18xlarge'|'ml.c5.24xlarge'|'ml.c5n.large'|'ml.c5n.2xlarge'|'ml.c5n.4xlarge'|'ml.c5n.9xlarge'|'ml.c5n.18xlarge'|'ml.m5.large'|'ml.m5.xlarge'|'ml.m5.2xlarge'|'ml.m5.4xlarge'|'ml.m5.8xlarge'|'ml.m5.12xlarge'|'ml.m5.16xlarge'|'ml.m5.24xlarge'|'ml.t3.medium'|'ml.t3.large'|'ml.t3.xlarge'|'ml.t3.2xlarge'|'ml.g6.xlarge'|'ml.g6.2xlarge'|'ml.g6.4xlarge'|'ml.g6.8xlarge'|'ml.g6.16xlarge'|'ml.g6.12xlarge'|'ml.g6.24xlarge'|'ml.g6.48xlarge'|'ml.gr6.4xlarge'|'ml.gr6.8xlarge'|'ml.g6e.xlarge'|'ml.g6e.2xlarge'|'ml.g6e.4xlarge'|'ml.g6e.8xlarge'|'ml.g6e.16xlarge'|'ml.g6e.12xlarge'|'ml.g6e.24xlarge'|'ml.g6e.48xlarge'|'ml.p5e.48xlarge'|'ml.p5en.48xlarge'|'ml.p6-b200.48xlarge'|'ml.trn2.3xlarge'|'ml.trn2.48xlarge'|'ml.c6i.large'|'ml.c6i.xlarge'|'ml.c6i.2xlarge'|'ml.c6i.4xlarge'|'ml.c6i.8xlarge'|'ml.c6i.12xlarge'|'ml.c6i.16xlarge'|'ml.c6i.24xlarge'|'ml.c6i.32xlarge'|'ml.m6i.large'|'ml.m6i.xlarge'|'ml.m6i.2xlarge'|'ml.m6i.4xlarge'|'ml.m6i.8xlarge'|'ml.m6i.12xlarge'|'ml.m6i.16xlarge'|'ml.m6i.24xlarge'|'ml.m6i.32xlarge'|'ml.r6i.large'|'ml.r6i.xlarge'|'ml.r6i.2xlarge'|'ml.r6i.4xlarge'|'ml.r6i.8xlarge'|'ml.r6i.12xlarge'|'ml.r6i.16xlarge'|'ml.r6i.24xlarge'|'ml.r6i.32xlarge'|'ml.i3en.large'|'ml.i3en.xlarge'|'ml.i3en.2xlarge'|'ml.i3en.3xlarge'|'ml.i3en.6xlarge'|'ml.i3en.12xlarge'|'ml.i3en.24xlarge'|'ml.m7i.large'|'ml.m7i.xlarge'|'ml.m7i.2xlarge'|'ml.m7i.4xlarge'|'ml.m7i.8xlarge'|'ml.m7i.12xlarge'|'ml.m7i.16xlarge'|'ml.m7i.24xlarge'|'ml.m7i.48xlarge'|'ml.r7i.large'|'ml.r7i.xlarge'|'ml.r7i.2xlarge'|'ml.r7i.4xlarge'|'ml.r7i.8xlarge'|'ml.r7i.12xlarge'|'ml.r7i.16xlarge'|'ml.r7i.24xlarge'|'ml.r7i.48xlarge',
'ExecutionRole': 'string',
'ThreadsPerCore': 123,
'InstanceStorageConfigs': [
{
'EbsVolumeConfig': {
'VolumeSizeInGB': 123,
'VolumeKmsKeyId': 'string',
'RootVolume': True|False
}
},
],
'OnStartDeepHealthChecks': [
'InstanceStress'|'InstanceConnectivity',
],
'TrainingPlanArn': 'string',
'OverrideVpcConfig': {
'SecurityGroupIds': [
'string',
],
'Subnets': [
'string',
]
},
'ScheduledUpdateConfig': {
'ScheduleExpression': 'string',
'DeploymentConfig': {
'RollingUpdatePolicy': {
'MaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
},
'RollbackMaximumBatchSize': {
'Type': 'INSTANCE_COUNT'|'CAPACITY_PERCENTAGE',
'Value': 123
}
},
'WaitIntervalInSeconds': 123,
'AutoRollbackConfiguration': [
{
'AlarmName': 'string'
},
]
}
},
'EnvironmentConfig': {
'FSxLustreConfig': {
'SizeInGiB': 123,
'PerUnitStorageThroughput': 123
}
}
},
],
TieredStorageConfig={
'Mode': 'Enable'|'Disable',
'InstanceMemoryAllocationPercentage': 123
},
NodeRecovery='Automatic'|'None',
InstanceGroupsToDelete=[
'string',
],
NodeProvisioningMode='Continuous',
ClusterRole='string',
AutoScaling={
'Mode': 'Enable'|'Disable',
'AutoScalerType': 'Karpenter'
}
)
string
[REQUIRED]
Specify the name of the SageMaker HyperPod cluster you want to update.
list
Specify the instance groups to update.
(dict) --
The specifications of an instance group that you need to define.
InstanceCount (integer) -- [REQUIRED]
Specifies the number of instances to add to the instance group of a SageMaker HyperPod cluster.
MinInstanceCount (integer) --
Defines the minimum number of instances required for an instance group to become InService. If this threshold isn't met within 3 hours, the instance group rolls back to its previous state - zero instances for new instance groups, or previous settings for existing instance groups. MinInstanceCount only affects the initial transition to InService and does not guarantee maintaining this minimum afterward.
InstanceGroupName (string) -- [REQUIRED]
Specifies the name of the instance group.
InstanceType (string) -- [REQUIRED]
Specifies the instance type of the instance group.
LifeCycleConfig (dict) -- [REQUIRED]
Specifies the LifeCycle configuration for the instance group.
SourceS3Uri (string) -- [REQUIRED]
An Amazon S3 bucket path where your lifecycle scripts are stored.
OnCreate (string) -- [REQUIRED]
The file name of the entrypoint script of lifecycle scripts under SourceS3Uri. This entrypoint script runs during cluster creation.
ExecutionRole (string) -- [REQUIRED]
Specifies an IAM execution role to be assumed by the instance group.
ThreadsPerCore (integer) --
Specifies the value for Threads per core. For instance types that support multithreading, you can specify 1 for disabling multithreading and 2 for enabling multithreading. For instance types that doesn't support multithreading, specify 1. For more information, see the reference table of CPU cores and threads per CPU core per instance type in the Amazon Elastic Compute Cloud User Guide.
InstanceStorageConfigs (list) --
Specifies the additional storage configurations for the instances in the SageMaker HyperPod cluster instance group.
(dict) --
Defines the configuration for attaching additional storage to the instances in the SageMaker HyperPod cluster instance group. To learn more, see SageMaker HyperPod release notes: June 20, 2024.
EbsVolumeConfig (dict) --
Defines the configuration for attaching additional Amazon Elastic Block Store (EBS) volumes to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeSizeInGB (integer) --
The size in gigabytes (GB) of the additional EBS volume to be attached to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeKmsKeyId (string) --
The ID of a KMS key to encrypt the Amazon EBS volume.
RootVolume (boolean) --
Specifies whether the configuration is for the cluster's root or secondary Amazon EBS volume. You can specify two ClusterEbsVolumeConfig fields to configure both the root and secondary volumes. Set the value to True if you'd like to provide your own customer managed Amazon Web Services KMS key to encrypt the root volume. When True:
The configuration is applied to the root volume.
You can't specify the VolumeSizeInGB field. The size of the root volume is determined for you.
You must specify a KMS key ID for VolumeKmsKeyId to encrypt the root volume with your own KMS key instead of an Amazon Web Services owned KMS key.
Otherwise, by default, the value is False, and the following applies:
The configuration is applied to the secondary volume, while the root volume is encrypted with an Amazon Web Services owned key.
You must specify the VolumeSizeInGB field.
You can optionally specify the VolumeKmsKeyId to encrypt the secondary volume with your own KMS key instead of an Amazon Web Services owned KMS key.
OnStartDeepHealthChecks (list) --
A flag indicating whether deep health checks should be performed when the cluster instance group is created or updated.
(string) --
TrainingPlanArn (string) --
The Amazon Resource Name (ARN); of the training plan to use for this cluster instance group.
For more information about how to reserve GPU capacity for your SageMaker HyperPod clusters using Amazon SageMaker Training Plan, see ``CreateTrainingPlan ``.
OverrideVpcConfig (dict) --
To configure multi-AZ deployments, customize the Amazon VPC configuration at the instance group level. You can specify different subnets and security groups across different AZs in the instance group specification to override a SageMaker HyperPod cluster's default Amazon VPC configuration. For more information about deploying a cluster in multiple AZs, see Setting up SageMaker HyperPod clusters across multiple AZs.
SecurityGroupIds (list) -- [REQUIRED]
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) -- [REQUIRED]
The ID of the subnets in the VPC to which you want to connect your training job or model. For information about the availability of specific instance types, see Supported Instance Types and Availability Zones.
(string) --
ScheduledUpdateConfig (dict) --
The configuration object of the schedule that SageMaker uses to update the AMI.
ScheduleExpression (string) -- [REQUIRED]
A cron expression that specifies the schedule that SageMaker follows when updating the AMI.
DeploymentConfig (dict) --
The configuration to use when updating the AMI versions.
RollingUpdatePolicy (dict) --
The policy that SageMaker uses when updating the AMI versions of the cluster.
MaximumBatchSize (dict) -- [REQUIRED]
The maximum amount of instances in the cluster that SageMaker can update at a time.
Type (string) -- [REQUIRED]
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) -- [REQUIRED]
Specifies the amount or percentage of instances SageMaker updates at a time.
RollbackMaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can roll back at a time.
Type (string) -- [REQUIRED]
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) -- [REQUIRED]
Specifies the amount or percentage of instances SageMaker updates at a time.
WaitIntervalInSeconds (integer) --
The duration in seconds that SageMaker waits before updating more instances in the cluster.
AutoRollbackConfiguration (list) --
An array that contains the alarms that SageMaker monitors to know whether to roll back the AMI update.
(dict) --
The details of the alarm to monitor during the AMI update.
AlarmName (string) -- [REQUIRED]
The name of the alarm.
ImageId (string) --
When configuring your HyperPod cluster, you can specify an image ID using one of the following options:
HyperPodPublicAmiId: Use a HyperPod public AMI
CustomAmiId: Use your custom AMI
default: Use the default latest system image
If you choose to use a custom AMI ( CustomAmiId), ensure it meets the following requirements:
Encryption: The custom AMI must be unencrypted.
Ownership: The custom AMI must be owned by the same Amazon Web Services account that is creating the HyperPod cluster.
Volume support: Only the primary AMI snapshot volume is supported; additional AMI volumes are not supported.
When updating the instance group's AMI through the UpdateClusterSoftware operation, if an instance group uses a custom AMI, you must provide an ImageId or use the default as input. Note that if you don't specify an instance group in your UpdateClusterSoftware request, then all of the instance groups are patched with the specified image.
KubernetesConfig (dict) --
Specifies the Kubernetes configuration for the instance group. You describe what you want the labels and taints to look like, and the cluster works to reconcile the actual state with the declared state for nodes in this instance group.
Labels (dict) --
Key-value pairs of labels to be applied to cluster nodes.
(string) --
(string) --
Taints (list) --
List of taints to be applied to cluster nodes.
(dict) --
A Kubernetes taint that can be applied to cluster nodes.
Key (string) -- [REQUIRED]
The key of the taint.
Value (string) --
The value of the taint.
Effect (string) -- [REQUIRED]
The effect of the taint. Valid values are NoSchedule, PreferNoSchedule, and NoExecute.
CapacityRequirements (dict) --
Specifies the capacity requirements for the instance group.
Spot (dict) --
Configuration options specific to Spot instances.
OnDemand (dict) --
Configuration options specific to On-Demand instances.
list
The specialized instance groups for training models like Amazon Nova to be created in the SageMaker HyperPod cluster.
(dict) --
The specifications of a restricted instance group that you need to define.
InstanceCount (integer) -- [REQUIRED]
Specifies the number of instances to add to the restricted instance group of a SageMaker HyperPod cluster.
InstanceGroupName (string) -- [REQUIRED]
Specifies the name of the restricted instance group.
InstanceType (string) -- [REQUIRED]
Specifies the instance type of the restricted instance group.
ExecutionRole (string) -- [REQUIRED]
Specifies an IAM execution role to be assumed by the restricted instance group.
ThreadsPerCore (integer) --
The number you specified to TreadsPerCore in CreateCluster for enabling or disabling multithreading. For instance types that support multithreading, you can specify 1 for disabling multithreading and 2 for enabling multithreading. For more information, see the reference table of CPU cores and threads per CPU core per instance type in the Amazon Elastic Compute Cloud User Guide.
InstanceStorageConfigs (list) --
Specifies the additional storage configurations for the instances in the SageMaker HyperPod cluster restricted instance group.
(dict) --
Defines the configuration for attaching additional storage to the instances in the SageMaker HyperPod cluster instance group. To learn more, see SageMaker HyperPod release notes: June 20, 2024.
EbsVolumeConfig (dict) --
Defines the configuration for attaching additional Amazon Elastic Block Store (EBS) volumes to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeSizeInGB (integer) --
The size in gigabytes (GB) of the additional EBS volume to be attached to the instances in the SageMaker HyperPod cluster instance group. The additional EBS volume is attached to each instance within the SageMaker HyperPod cluster instance group and mounted to /opt/sagemaker.
VolumeKmsKeyId (string) --
The ID of a KMS key to encrypt the Amazon EBS volume.
RootVolume (boolean) --
Specifies whether the configuration is for the cluster's root or secondary Amazon EBS volume. You can specify two ClusterEbsVolumeConfig fields to configure both the root and secondary volumes. Set the value to True if you'd like to provide your own customer managed Amazon Web Services KMS key to encrypt the root volume. When True:
The configuration is applied to the root volume.
You can't specify the VolumeSizeInGB field. The size of the root volume is determined for you.
You must specify a KMS key ID for VolumeKmsKeyId to encrypt the root volume with your own KMS key instead of an Amazon Web Services owned KMS key.
Otherwise, by default, the value is False, and the following applies:
The configuration is applied to the secondary volume, while the root volume is encrypted with an Amazon Web Services owned key.
You must specify the VolumeSizeInGB field.
You can optionally specify the VolumeKmsKeyId to encrypt the secondary volume with your own KMS key instead of an Amazon Web Services owned KMS key.
OnStartDeepHealthChecks (list) --
A flag indicating whether deep health checks should be performed when the cluster restricted instance group is created or updated.
(string) --
TrainingPlanArn (string) --
The Amazon Resource Name (ARN) of the training plan to filter clusters by. For more information about reserving GPU capacity for your SageMaker HyperPod clusters using Amazon SageMaker Training Plan, see ``CreateTrainingPlan ``.
OverrideVpcConfig (dict) --
Specifies an Amazon Virtual Private Cloud (VPC) that your SageMaker jobs, hosted models, and compute resources have access to. You can control access to and from your resources by configuring a VPC. For more information, see Give SageMaker Access to Resources in your Amazon VPC.
SecurityGroupIds (list) -- [REQUIRED]
The VPC security group IDs, in the form sg-xxxxxxxx. Specify the security groups for the VPC that is specified in the Subnets field.
(string) --
Subnets (list) -- [REQUIRED]
The ID of the subnets in the VPC to which you want to connect your training job or model. For information about the availability of specific instance types, see Supported Instance Types and Availability Zones.
(string) --
ScheduledUpdateConfig (dict) --
The configuration object of the schedule that SageMaker follows when updating the AMI.
ScheduleExpression (string) -- [REQUIRED]
A cron expression that specifies the schedule that SageMaker follows when updating the AMI.
DeploymentConfig (dict) --
The configuration to use when updating the AMI versions.
RollingUpdatePolicy (dict) --
The policy that SageMaker uses when updating the AMI versions of the cluster.
MaximumBatchSize (dict) -- [REQUIRED]
The maximum amount of instances in the cluster that SageMaker can update at a time.
Type (string) -- [REQUIRED]
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) -- [REQUIRED]
Specifies the amount or percentage of instances SageMaker updates at a time.
RollbackMaximumBatchSize (dict) --
The maximum amount of instances in the cluster that SageMaker can roll back at a time.
Type (string) -- [REQUIRED]
Specifies whether SageMaker should process the update by amount or percentage of instances.
Value (integer) -- [REQUIRED]
Specifies the amount or percentage of instances SageMaker updates at a time.
WaitIntervalInSeconds (integer) --
The duration in seconds that SageMaker waits before updating more instances in the cluster.
AutoRollbackConfiguration (list) --
An array that contains the alarms that SageMaker monitors to know whether to roll back the AMI update.
(dict) --
The details of the alarm to monitor during the AMI update.
AlarmName (string) -- [REQUIRED]
The name of the alarm.
EnvironmentConfig (dict) -- [REQUIRED]
The configuration for the restricted instance groups (RIG) environment.
FSxLustreConfig (dict) --
Configuration settings for an Amazon FSx for Lustre file system to be used with the cluster.
SizeInGiB (integer) -- [REQUIRED]
The storage capacity of the Amazon FSx for Lustre file system, specified in gibibytes (GiB).
PerUnitStorageThroughput (integer) -- [REQUIRED]
The throughput capacity of the Amazon FSx for Lustre file system, measured in MB/s per TiB of storage.
dict
Updates the configuration for managed tier checkpointing on the HyperPod cluster. For example, you can enable or disable the feature and modify the percentage of cluster memory allocated for checkpoint storage.
Mode (string) -- [REQUIRED]
Specifies whether managed tier checkpointing is enabled or disabled for the HyperPod cluster. When set to Enable, the system installs a memory management daemon that provides disaggregated memory as a service for checkpoint storage. When set to Disable, the feature is turned off and the memory management daemon is removed from the cluster.
InstanceMemoryAllocationPercentage (integer) --
The percentage (int) of cluster memory to allocate for checkpointing.
string
The node recovery mode to be applied to the SageMaker HyperPod cluster.
list
Specify the names of the instance groups to delete. Use a single , as the separator between multiple names.
(string) --
string
Determines how instance provisioning is handled during cluster operations. In Continuous mode, the cluster provisions available instances incrementally and retries until the target count is reached. The cluster becomes operational once cluster-level resources are ready. Use CurrentCount and TargetCount in DescribeCluster to track provisioning progress.
string
The Amazon Resource Name (ARN) of the IAM role that HyperPod assumes for cluster autoscaling operations. Cannot be updated while autoscaling is enabled.
dict
Updates the autoscaling configuration for the cluster. Use to enable or disable automatic node scaling.
Mode (string) -- [REQUIRED]
Describes whether autoscaling is enabled or disabled for the cluster. Valid values are Enable and Disable.
AutoScalerType (string) --
The type of autoscaler to use. Currently supported value is Karpenter.
dict
Response Syntax
{
'ClusterArn': 'string'
}
Response Structure
(dict) --
ClusterArn (string) --
The Amazon Resource Name (ARN) of the updated SageMaker HyperPod cluster.