/AWS1/CL_SGMCLUSTINSTGROUPDETS¶
Details of an instance group in a SageMaker HyperPod cluster.
CONSTRUCTOR¶
IMPORTING¶
Optional arguments:¶
iv_currentcount TYPE /AWS1/SGMCLUSTNONNEGINSTCOUNT /AWS1/SGMCLUSTNONNEGINSTCOUNT¶
The number of instances that are currently in the instance group of a SageMaker HyperPod cluster.
iv_targetcount TYPE /AWS1/SGMCLUSTERINSTANCECOUNT /AWS1/SGMCLUSTERINSTANCECOUNT¶
The number of instances you specified to add to the instance group of a SageMaker HyperPod cluster.
iv_mincount TYPE /AWS1/SGMCLUSTERINSTANCECOUNT /AWS1/SGMCLUSTERINSTANCECOUNT¶
The minimum number of instances that must be available in the instance group of a SageMaker HyperPod cluster before it transitions to
InServicestatus.
iv_instancegroupname TYPE /AWS1/SGMCLUSTERINSTGROUPNAME /AWS1/SGMCLUSTERINSTGROUPNAME¶
The name of the instance group of a SageMaker HyperPod cluster.
iv_instancetype TYPE /AWS1/SGMCLUSTERINSTANCETYPE /AWS1/SGMCLUSTERINSTANCETYPE¶
The instance type of the instance group of a SageMaker HyperPod cluster.
io_instancerequirements TYPE REF TO /AWS1/CL_SGMCLUSTINSTREQDETS /AWS1/CL_SGMCLUSTINSTREQDETS¶
The instance requirements for the instance group, including the current and desired instance types. This field is present for flexible instance groups that support multiple instance types.
it_instancetypedetails TYPE /AWS1/CL_SGMCLUSTINSTTYPEDET=>TT_CLUSTERINSTANCETYPEDETAILS TT_CLUSTERINSTANCETYPEDETAILS¶
Details about the instance types in the instance group, including the count and configuration of each instance type. This field is present for flexible instance groups that support multiple instance types.
io_lifecycleconfig TYPE REF TO /AWS1/CL_SGMCLUSTERLCCONFIG /AWS1/CL_SGMCLUSTERLCCONFIG¶
Details of LifeCycle configuration for the instance group.
iv_executionrole TYPE /AWS1/SGMROLEARN /AWS1/SGMROLEARN¶
The execution role for the instance group to assume.
iv_threadspercore TYPE /AWS1/SGMCLUSTERTHREADSPERCORE /AWS1/SGMCLUSTERTHREADSPERCORE¶
The number you specified to
TreadsPerCoreinCreateClusterfor 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.
it_instancestorageconfigs TYPE /AWS1/CL_SGMCLUSTINSTSTRGCFG=>TT_CLUSTERINSTSTORAGECONFIGS TT_CLUSTERINSTSTORAGECONFIGS¶
The additional storage configurations for the instances in the SageMaker HyperPod cluster instance group.
it_onstartdeephealthchecks TYPE /AWS1/CL_SGMONSTRTDEEPHEALTH00=>TT_ONSTARTDEEPHEALTHCHECKS TT_ONSTARTDEEPHEALTHCHECKS¶
A flag indicating whether deep health checks should be performed when the cluster instance group is created or updated.
iv_status TYPE /AWS1/SGMINSTANCEGROUPSTATUS /AWS1/SGMINSTANCEGROUPSTATUS¶
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.
iv_trainingplanarn TYPE /AWS1/SGMTRAININGPLANARN /AWS1/SGMTRAININGPLANARN¶
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.
iv_trainingplanstatus TYPE /AWS1/SGMINSTGROUPTRNPLANSTAT /AWS1/SGMINSTGROUPTRNPLANSTAT¶
The current status of the training plan associated with this cluster instance group.
io_overridevpcconfig TYPE REF TO /AWS1/CL_SGMVPCCONFIG /AWS1/CL_SGMVPCCONFIG¶
The customized Amazon VPC configuration at the instance group level that overrides the default Amazon VPC configuration of the SageMaker HyperPod cluster.
io_scheduledupdateconfig TYPE REF TO /AWS1/CL_SGMSCHDDUPDATECONFIG /AWS1/CL_SGMSCHDDUPDATECONFIG¶
The configuration object of the schedule that SageMaker follows when updating the AMI.
iv_currentimageid TYPE /AWS1/SGMIMAGEID /AWS1/SGMIMAGEID¶
The ID of the Amazon Machine Image (AMI) currently in use by the instance group.
iv_desiredimageid TYPE /AWS1/SGMIMAGEID /AWS1/SGMIMAGEID¶
The ID of the Amazon Machine Image (AMI) desired for the instance group.
it_activeoperations TYPE /AWS1/CL_SGMACTIVEOPERATIONS_W=>TT_ACTIVEOPERATIONS TT_ACTIVEOPERATIONS¶
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
Scalingwith value 1.
io_kubernetesconfig TYPE REF TO /AWS1/CL_SGMCLUSTK8SCONFIGDETS /AWS1/CL_SGMCLUSTK8SCONFIGDETS¶
The Kubernetes configuration for the instance group that contains labels and taints to be applied for the nodes in this instance group.
io_capacityrequirements TYPE REF TO /AWS1/CL_SGMCLUSTERCAPREQS /AWS1/CL_SGMCLUSTERCAPREQS¶
The instance capacity requirements for the instance group.
iv_targetstatecount TYPE /AWS1/SGMCLUSTERINSTANCECOUNT /AWS1/SGMCLUSTERINSTANCECOUNT¶
Represents the number of running nodes using the desired Image ID.
During software update operations: This count shows the number of nodes running on the desired Image ID. If a rollback occurs, the current image ID and desired image ID (both included in the describe cluster response) swap values. The TargetStateCount then shows the number of nodes running on the newly designated desired image ID (which was previously the current image ID).
During simultaneous scaling and software update operations: This count shows the number of instances running on the desired image ID, including any new instances created as part of the scaling request. New nodes are always created using the desired image ID, so TargetStateCount reflects the total count of nodes running on the desired image ID, even during rollback scenarios.
iv_softwareupdatestatus TYPE /AWS1/SGMSOFTWAREUPDATESTATUS /AWS1/SGMSOFTWAREUPDATESTATUS¶
Status of the last software udpate request.
Status transitions follow these possible sequences:
Pending -> InProgress -> Succeeded
Pending -> InProgress -> RollbackInProgress -> RollbackComplete
Pending -> InProgress -> RollbackInProgress -> Failed
io_actsoftwareupdateconfig TYPE REF TO /AWS1/CL_SGMDEPLOYMENTCONF /AWS1/CL_SGMDEPLOYMENTCONF¶
The configuration to use when updating the AMI versions.
io_slurmconfig TYPE REF TO /AWS1/CL_SGMCLUSTSLURMCFGDETS /AWS1/CL_SGMCLUSTSLURMCFGDETS¶
The Slurm configuration for the instance group.
io_networkinterface TYPE REF TO /AWS1/CL_SGMCLUSTNETIFACEDETS /AWS1/CL_SGMCLUSTNETIFACEDETS¶
The network interface configuration for the instance group.
Queryable Attributes¶
CurrentCount¶
The number of instances that are currently in the instance group of a SageMaker HyperPod cluster.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_CURRENTCOUNT() |
Getter for CURRENTCOUNT, with configurable default |
ASK_CURRENTCOUNT() |
Getter for CURRENTCOUNT w/ exceptions if field has no value |
HAS_CURRENTCOUNT() |
Determine if CURRENTCOUNT has a value |
TargetCount¶
The number of instances you specified to add to the instance group of a SageMaker HyperPod cluster.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_TARGETCOUNT() |
Getter for TARGETCOUNT, with configurable default |
ASK_TARGETCOUNT() |
Getter for TARGETCOUNT w/ exceptions if field has no value |
HAS_TARGETCOUNT() |
Determine if TARGETCOUNT has a value |
MinCount¶
The minimum number of instances that must be available in the instance group of a SageMaker HyperPod cluster before it transitions to
InServicestatus.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_MINCOUNT() |
Getter for MINCOUNT, with configurable default |
ASK_MINCOUNT() |
Getter for MINCOUNT w/ exceptions if field has no value |
HAS_MINCOUNT() |
Determine if MINCOUNT has a value |
InstanceGroupName¶
The name of the instance group of a SageMaker HyperPod cluster.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_INSTANCEGROUPNAME() |
Getter for INSTANCEGROUPNAME, with configurable default |
ASK_INSTANCEGROUPNAME() |
Getter for INSTANCEGROUPNAME w/ exceptions if field has no v |
HAS_INSTANCEGROUPNAME() |
Determine if INSTANCEGROUPNAME has a value |
InstanceType¶
The instance type of the instance group of a SageMaker HyperPod cluster.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_INSTANCETYPE() |
Getter for INSTANCETYPE, with configurable default |
ASK_INSTANCETYPE() |
Getter for INSTANCETYPE w/ exceptions if field has no value |
HAS_INSTANCETYPE() |
Determine if INSTANCETYPE has a value |
InstanceRequirements¶
The instance requirements for the instance group, including the current and desired instance types. This field is present for flexible instance groups that support multiple instance types.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_INSTANCEREQUIREMENTS() |
Getter for INSTANCEREQUIREMENTS |
InstanceTypeDetails¶
Details about the instance types in the instance group, including the count and configuration of each instance type. This field is present for flexible instance groups that support multiple instance types.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_INSTANCETYPEDETAILS() |
Getter for INSTANCETYPEDETAILS, with configurable default |
ASK_INSTANCETYPEDETAILS() |
Getter for INSTANCETYPEDETAILS w/ exceptions if field has no |
HAS_INSTANCETYPEDETAILS() |
Determine if INSTANCETYPEDETAILS has a value |
LifeCycleConfig¶
Details of LifeCycle configuration for the instance group.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_LIFECYCLECONFIG() |
Getter for LIFECYCLECONFIG |
ExecutionRole¶
The execution role for the instance group to assume.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_EXECUTIONROLE() |
Getter for EXECUTIONROLE, with configurable default |
ASK_EXECUTIONROLE() |
Getter for EXECUTIONROLE w/ exceptions if field has no value |
HAS_EXECUTIONROLE() |
Determine if EXECUTIONROLE has a value |
ThreadsPerCore¶
The number you specified to
TreadsPerCoreinCreateClusterfor 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.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_THREADSPERCORE() |
Getter for THREADSPERCORE, with configurable default |
ASK_THREADSPERCORE() |
Getter for THREADSPERCORE w/ exceptions if field has no valu |
HAS_THREADSPERCORE() |
Determine if THREADSPERCORE has a value |
InstanceStorageConfigs¶
The additional storage configurations for the instances in the SageMaker HyperPod cluster instance group.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_INSTANCESTORAGECONFIGS() |
Getter for INSTANCESTORAGECONFIGS, with configurable default |
ASK_INSTANCESTORAGECONFIGS() |
Getter for INSTANCESTORAGECONFIGS w/ exceptions if field has |
HAS_INSTANCESTORAGECONFIGS() |
Determine if INSTANCESTORAGECONFIGS has a value |
OnStartDeepHealthChecks¶
A flag indicating whether deep health checks should be performed when the cluster instance group is created or updated.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_ONSTARTDEEPHEALTHCHECKS() |
Getter for ONSTARTDEEPHEALTHCHECKS, with configurable defaul |
ASK_ONSTARTDEEPHEALTHCHECKS() |
Getter for ONSTARTDEEPHEALTHCHECKS w/ exceptions if field ha |
HAS_ONSTARTDEEPHEALTHCHECKS() |
Determine if ONSTARTDEEPHEALTHCHECKS has a value |
Status¶
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.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_STATUS() |
Getter for STATUS, with configurable default |
ASK_STATUS() |
Getter for STATUS w/ exceptions if field has no value |
HAS_STATUS() |
Determine if STATUS has a value |
TrainingPlanArn¶
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.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_TRAININGPLANARN() |
Getter for TRAININGPLANARN, with configurable default |
ASK_TRAININGPLANARN() |
Getter for TRAININGPLANARN w/ exceptions if field has no val |
HAS_TRAININGPLANARN() |
Determine if TRAININGPLANARN has a value |
TrainingPlanStatus¶
The current status of the training plan associated with this cluster instance group.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_TRAININGPLANSTATUS() |
Getter for TRAININGPLANSTATUS, with configurable default |
ASK_TRAININGPLANSTATUS() |
Getter for TRAININGPLANSTATUS w/ exceptions if field has no |
HAS_TRAININGPLANSTATUS() |
Determine if TRAININGPLANSTATUS has a value |
OverrideVpcConfig¶
The customized Amazon VPC configuration at the instance group level that overrides the default Amazon VPC configuration of the SageMaker HyperPod cluster.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_OVERRIDEVPCCONFIG() |
Getter for OVERRIDEVPCCONFIG |
ScheduledUpdateConfig¶
The configuration object of the schedule that SageMaker follows when updating the AMI.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_SCHEDULEDUPDATECONFIG() |
Getter for SCHEDULEDUPDATECONFIG |
CurrentImageId¶
The ID of the Amazon Machine Image (AMI) currently in use by the instance group.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_CURRENTIMAGEID() |
Getter for CURRENTIMAGEID, with configurable default |
ASK_CURRENTIMAGEID() |
Getter for CURRENTIMAGEID w/ exceptions if field has no valu |
HAS_CURRENTIMAGEID() |
Determine if CURRENTIMAGEID has a value |
DesiredImageId¶
The ID of the Amazon Machine Image (AMI) desired for the instance group.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_DESIREDIMAGEID() |
Getter for DESIREDIMAGEID, with configurable default |
ASK_DESIREDIMAGEID() |
Getter for DESIREDIMAGEID w/ exceptions if field has no valu |
HAS_DESIREDIMAGEID() |
Determine if DESIREDIMAGEID has a value |
ActiveOperations¶
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
Scalingwith value 1.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_ACTIVEOPERATIONS() |
Getter for ACTIVEOPERATIONS, with configurable default |
ASK_ACTIVEOPERATIONS() |
Getter for ACTIVEOPERATIONS w/ exceptions if field has no va |
HAS_ACTIVEOPERATIONS() |
Determine if ACTIVEOPERATIONS has a value |
KubernetesConfig¶
The Kubernetes configuration for the instance group that contains labels and taints to be applied for the nodes in this instance group.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_KUBERNETESCONFIG() |
Getter for KUBERNETESCONFIG |
CapacityRequirements¶
The instance capacity requirements for the instance group.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_CAPACITYREQUIREMENTS() |
Getter for CAPACITYREQUIREMENTS |
TargetStateCount¶
Represents the number of running nodes using the desired Image ID.
During software update operations: This count shows the number of nodes running on the desired Image ID. If a rollback occurs, the current image ID and desired image ID (both included in the describe cluster response) swap values. The TargetStateCount then shows the number of nodes running on the newly designated desired image ID (which was previously the current image ID).
During simultaneous scaling and software update operations: This count shows the number of instances running on the desired image ID, including any new instances created as part of the scaling request. New nodes are always created using the desired image ID, so TargetStateCount reflects the total count of nodes running on the desired image ID, even during rollback scenarios.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_TARGETSTATECOUNT() |
Getter for TARGETSTATECOUNT, with configurable default |
ASK_TARGETSTATECOUNT() |
Getter for TARGETSTATECOUNT w/ exceptions if field has no va |
HAS_TARGETSTATECOUNT() |
Determine if TARGETSTATECOUNT has a value |
SoftwareUpdateStatus¶
Status of the last software udpate request.
Status transitions follow these possible sequences:
Pending -> InProgress -> Succeeded
Pending -> InProgress -> RollbackInProgress -> RollbackComplete
Pending -> InProgress -> RollbackInProgress -> Failed
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_SOFTWAREUPDATESTATUS() |
Getter for SOFTWAREUPDATESTATUS, with configurable default |
ASK_SOFTWAREUPDATESTATUS() |
Getter for SOFTWAREUPDATESTATUS w/ exceptions if field has n |
HAS_SOFTWAREUPDATESTATUS() |
Determine if SOFTWAREUPDATESTATUS has a value |
ActiveSoftwareUpdateConfig¶
The configuration to use when updating the AMI versions.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_ACTSOFTWAREUPDATECONFIG() |
Getter for ACTIVESOFTWAREUPDATECONFIG |
SlurmConfig¶
The Slurm configuration for the instance group.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_SLURMCONFIG() |
Getter for SLURMCONFIG |
NetworkInterface¶
The network interface configuration for the instance group.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_NETWORKINTERFACE() |
Getter for NETWORKINTERFACE |
Public Local Types In This Class¶
Internal table types, representing arrays and maps of this class, are defined as local types:
TT_CLUSTERINSTGROUPDETAILSLIST¶
TYPES TT_CLUSTERINSTGROUPDETAILSLIST TYPE STANDARD TABLE OF REF TO /AWS1/CL_SGMCLUSTINSTGROUPDETS WITH DEFAULT KEY
.