Get node groups | NVIDIA Fleet Intelligence Documentation

{
  "hasMore": false,
  "nodeGroups": [
    {
      "alertOverview": {
        "criticalCount": 2,
        "infoCount": 15,
        "lastCriticalAlert": "2024-01-15T14:15:00Z",
        "lastWarningAlert": "2024-01-15T14:28:00Z",
        "totalCount": 24,
        "unknownCount": 15,
        "warningCount": 7
      },
      "computeZoneContact": {
        "email": "admin@example.com",
        "pic": "John Doe"
      },
      "computeZoneGeoLocation": {
        "city": "Baltimore",
        "country": "United States",
        "latitude": 39.0458,
        "longitude": -76.6413,
        "region": "us-east-1"
      },
      "computeZoneId": "cz-us-east-1",
      "computeZoneName": "US East 1 ComputeZone",
      "cpuCoresCount": 40,
      "createdAt": "2024-01-10T09:00:00Z",
      "degradedNodesCount": 2,
      "description": "High-performance GPU nodes for ML workloads",
      "gpusCount": 80,
      "healthPercentage": 95,
      "healthState": "Healthy",
      "healthyNodesCount": 38,
      "id": "ng-gpu-cluster-1",
      "metrics": [
        {
          "aggregation": "average",
          "description": "System memory metrics",
          "lastUpdated": "2024-01-15T14:30:00Z",
          "name": "memory",
          "unit": "percent",
          "value": 72.3
        }
      ],
      "name": "GPU Compute Cluster 1",
      "nodesCount": 40,
      "tags": [
        "ml-training",
        "gpu-intensive",
        "high-memory"
      ],
      "unhealthyNodesCount": 0,
      "unknownNodesCount": 0,
      "updatedAt": "2024-01-15T14:28:00Z"
    }
  ],
  "page": 0,
  "pageSize": 20,
  "total": 0
}

Get list of node groups with optional filtering and different view modes. Use view=detail for comprehensive metrics and view=basic for minimal ID/Name list. Detail view supports additional filters: computeZoneNames (partial match), healthStatuses, and gpuTypes

Query parameters

viewenumOptionalDefaults to detail

View mode: ‘detail’ for comprehensive metrics (default) or ‘basic’ for minimal list

Allowed values:

includeMetricsbooleanOptionalDefaults to true

[DETAIL VIEW ONLY] Include metrics in response (default: true). Set to false for better performance when metrics are not needed.

computeZoneIdslist of stringsOptional

Filter by compute zone IDs (e.g., ?computeZoneIds=id1&computeZoneIds=id2)

computeZoneNameslist of stringsOptional

[DETAIL VIEW ONLY] Filter by compute zone names with partial match (e.g., ?computeZoneNames=east&computeZoneNames=west)

computeZoneIdstringOptional

[LEGACY] Single compute zone ID to filter node groups - use computeZoneIds instead

nodeGroupIdslist of stringsOptional

List of node group IDs to filter (e.g., ?nodeGroupIds=id1&nodeGroupIds=id2)

healthStatuseslist of enumsOptional

[DETAIL VIEW ONLY] Filter by node health status values (e.g., ?healthStatuses=Healthy&healthStatuses=Degraded)

Allowed values:

gpuTypeslist of stringsOptional

[DETAIL VIEW ONLY] Filter by GPU type values (e.g., ?gpuTypes=NVIDIA-H100-80GB-HBM3)

sortByenumOptionalDefaults to health

Field to sort by (health, nodes)

Allowed values:

orderenumOptionalDefaults to desc

Sort order (asc, desc)

Allowed values:

pageintegerOptionalDefaults to 0

Page number (default: 0)

pageSizeintegerOptionalDefaults to 20

Page size (default: 20, max: 100)

Response

Basic view response (view=basic)

models.NodeGroupsResponseobject

models.ListNodeGroupsResponseobject

Errors

400

Bad Request Error

401

Unauthorized Error

403

Forbidden Error

499

Client Closed Request Error

500

Internal Server Error

503

Service Unavailable Error