AWS pcs documentation change
Summary
Added 'Scale-down idle time' configuration option for compute node groups and renumbered steps. Included CLI example for setting scaleDownIdleTimeInSeconds.
Security assessment
Change introduces operational optimization feature (node scaling) without addressing vulnerabilities. No security fixes, patches, or vulnerability mitigations are mentioned.
Diff
diff --git a/pcs/latest/userguide/working-with_cng_create.md b/pcs/latest/userguide/working-with_cng_create.md index 67d39e232..8a5f83e01 100644 --- a//pcs/latest/userguide/working-with_cng_create.md +++ b//pcs/latest/userguide/working-with_cng_create.md @@ -77 +77 @@ AWS PCS creates a managed launch template for each compute node group. These are - 6. (Optional) In the **Slurm custom settings** section, you can add parameter name and value pairs to configure additional Slurm settings. For a complete list of supported parameters, see [Custom Slurm settings for AWS PCS compute node groups](./slurm-custom-settings-cng.html). + 6. (Optional) In the **Scheduler configuration** section, you can specify the following: @@ -79 +79 @@ AWS PCS creates a managed launch template for each compute node group. These are - 7. (Optional) Under **Tags** , add any tags to your compute node group. + 1. **Scale-down idle time** – (Optional) The time in seconds before an idle node in this node group is scaled down. If not set, the cluster-level value is used. This setting requires Slurm version 25.11 or later. @@ -81 +81,5 @@ AWS PCS creates a managed launch template for each compute node group. These are - 8. Choose **Create compute node group**. The **Status** field shows `Creating` while AWS PCS provisions the node group. This can take several minutes. + 7. (Optional) In the **Slurm custom settings** section, you can add parameter name and value pairs to configure additional Slurm settings. For a complete list of supported parameters, see [Custom Slurm settings for AWS PCS compute node groups](./slurm-custom-settings-cng.html). + + 8. (Optional) Under **Tags** , add any tags to your compute node group. + + 9. Choose **Create compute node group**. The **Status** field shows `Creating` while AWS PCS provisions the node group. This can take several minutes. @@ -150,0 +155,15 @@ For more information, see [Custom Slurm settings for AWS PCS compute node groups +###### Example– Creating a compute node group with a scale-down idle time + + + aws pcs create-compute-node-group --region region \ + --cluster-identifier my-cluster \ + --compute-node-group-name my-gpu-nodes \ + --subnet-ids subnet-ExampleID1 \ + --custom-launch-template id=lt-ExampleID1,version='1' \ + --iam-instance-profile-arn=arn:InstanceProfile \ + --scaling-config minInstanceCount=0,maxInstanceCount=10 \ + --instance-configs instanceType=p4d.24xlarge \ + --slurm-configuration scaleDownIdleTimeInSeconds=300 + +The `scaleDownIdleTimeInSeconds` setting at the compute node group level overrides the cluster-level value for nodes in this group. This setting requires Slurm version 25.11 or later. +