Deleting a compute node group in AWS PCS
This topic provides an overview of available options and describes what to consider when you delete an compute node group in AWS PCS.
Considerations when deleting a
compute node group
Compute node groups define EC2 instances that are used to process jobs, provide interactive shell access, and other tasks. They are often associated with one or more AWS PCS queues. Before you delete a compute node group, consider the following:
-
Any EC2 instances launched by the compute node group will be terminated. This will cancel jobs that are running on these instances, and terminate running interactive processes.
-
You must disassociate the compute node group from all queues before you can delete it. For more information, see Updating an AWS PCS queue.
Delete the compute node group
You can use the AWS Management Console or AWS CLI to delete a compute node group.
- AWS Management Console
-
To delete a compute node group
-
Open the AWS PCS
console.
-
Select the cluster of the compute node group.
-
Navigate to Compute node groups and select the compute node group to delete.
-
Choose Delete.
-
The Status field shows Deleting. It can take several
minutes to complete.
You can use commands native to your scheduler to confirm that the compute node group is deleted. For
example, use sinfo or squeue for Slurm.
- AWS CLI
-
To delete a compute node group
-
Use the following command to delete a compute node group, with these replacements:
-
Replace region-code with the AWS Region your cluster is
in.
-
Replace my-node-group with the name or ID of your compute node group.
-
Replace my-cluster with the name or ID of your cluster.
aws pcs delete-compute-node-group --region region-code \
--compute-node-group-identifier my-node-group \
--cluster-identifier my-cluster
It can take several minutes to delete the compute node group.
You can use commands native to your scheduler to confirm that the compute node group is deleted. For
example, use sinfo or squeue for Slurm.
