Add UltraServer support for CloudWatch agent

[petruanica]

Description of the issue

• Add support for monitoring EC2 UltraServers using CloudWatch Agent

Description of changes

• Allowlisted UltraServer dimension for Neuron metrics
• Added new (ClusterName, UltraServer) dimension for Neuron metrics emitted at the node level

License

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

Tests

• Updated test files to include the new UltraServer dimension
• Verified metric output format includes the new UltraServer identifier by deploying changes in EKS test cluster

EMF sample:

{
    "CloudWatchMetrics": [
        {
            "Namespace": "ContainerInsights",
            "Dimensions": [
                [
                    "ClusterName"
                ],
                [
                    "ClusterName",
                    "UltraServer"
                ],
                [
                    "ClusterName",
                    "InstanceId",
                    "NodeName"
                ],
                [
                    "ClusterName",
                    "InstanceId",
                    "InstanceType",
                    "NeuronCore",
                    "NeuronDevice",
                    "NodeName"
                ]
            ],
            "Metrics": [
                {
                    "Name": "node_neuroncore_memory_usage_total",
                    "Unit": "Bytes",
                    "StorageResolution": 60
                },
                {
                    "Name": "node_neuroncore_utilization",
                    "Unit": "Percent",
                    "StorageResolution": 60
                },
                {
                    "Name": "node_neuroncore_memory_usage_runtime_memory",
                    "Unit": "Bytes",
                    "StorageResolution": 60
                },
                {
                    "Name": "node_neuroncore_memory_usage_constants",
                    "Unit": "Bytes",
                    "StorageResolution": 60
                },
                {
                    "Name": "node_neuroncore_memory_usage_model_code",
                    "Unit": "Bytes",
                    "StorageResolution": 60
                },
                {
                    "Name": "node_neuroncore_memory_usage_model_shared_scratchpad",
                    "Unit": "Bytes",
                    "StorageResolution": 60
                },
                {
                    "Name": "node_neuroncore_memory_usage_tensors",
                    "Unit": "Bytes",
                    "StorageResolution": 60
                }
            ]
        }
    ],
    "ClusterName": "HyperPodNeuronEFA",
    "InstanceId": "i-07c72013a77a6826e",
    "InstanceType": "trn1.2xlarge",
    "NeuronCore": "core1",
    "NeuronDevice": "device0",
    "NodeName": "ip-192-168-94-194.us-west-2.compute.internal",
    "Timestamp": "1740591416636",
    "Type": "NodeAWSNeuronCore",
    "UltraServer": "u-1234567890",
    "Version": "0",
    "availability_zone": "us-west-2b",
    "kubernetes": {
        "host": "ip-192-168-94-194.us-west-2.compute.internal"
    },
    "region": "us-west-2",
    "subnet_id": "subnet-0dfa65d0c9792b6f3",
    "node_neuroncore_memory_usage_constants": 0,
    "node_neuroncore_memory_usage_model_code": 0,
    "node_neuroncore_memory_usage_model_shared_scratchpad": 0,
    "node_neuroncore_memory_usage_runtime_memory": 0,
    "node_neuroncore_memory_usage_tensors": 0,
    "node_neuroncore_memory_usage_total": 0,
    "node_neuroncore_utilization": 0
}

Requirements

Before commit the code, please do the following steps.

1. Run make fmt and make fmt-sh
2. Run make lint

[github-actions]

This PR was marked stale due to lack of activity.