Get deployment metrics

Returns per-node GPU/CPU/memory utilization and Knative queue-proxy request rate / concurrency / latency for the trainer pods. The sampler half of a Loops deployment is an OracleVersion and uses the existing model-metrics endpoint.

POST /v1/loops/deployments/{deployment_id}/metrics

cURL

cURL

curl --request POST \
--url https://api.baseten.co/v1/loops/deployments/{deployment_id}/metrics \
--header "Authorization: Bearer $BASETEN_API_KEY" \
--data '{
  "end_epoch_millis": null,
  "start_epoch_millis": null,
  "step_seconds": null,
  "time_divisor_seconds": null
}'

200

{
  "deployment_id": "<string>",
  "metrics": {
    "inference_volume": [
      {
        "value": 123,
        "timestamp": "2023-11-07T05:31:56Z"
      }
    ],
    "concurrent_requests": [
      {
        "value": 123,
        "timestamp": "2023-11-07T05:31:56Z"
      }
    ],
    "response_time_stats": [
      {
        "timestamp": "2023-11-07T05:31:56Z",
        "p50": 123,
        "p95": 123,
        "p99": 123
      }
    ],
    "inference_volume_by_status": [
      {
        "timestamp": "2023-11-07T05:31:56Z",
        "status_2xx": 123,
        "status_4xx": 123,
        "status_5xx": 123
      }
    ],
    "gpu_memory_usage_bytes": {},
    "gpu_utilization": {},
    "cpu_usage": [
      {
        "value": 123,
        "timestamp": "2023-11-07T05:31:56Z"
      }
    ],
    "cpu_memory_usage_bytes": [
      {
        "value": 123,
        "timestamp": "2023-11-07T05:31:56Z"
      }
    ],
    "ephemeral_storage": {
      "usage_bytes": [
        {
          "value": 123,
          "timestamp": "2023-11-07T05:31:56Z"
        }
      ],
      "utilization": [
        {
          "value": 123,
          "timestamp": "2023-11-07T05:31:56Z"
        }
      ]
    },
    "per_node_metrics": [
      {
        "node_id": "<string>",
        "gpu_memory_usage_bytes": {},
        "gpu_utilization": {},
        "cpu_usage": [
          {
            "value": 123,
            "timestamp": "2023-11-07T05:31:56Z"
          }
        ],
        "cpu_memory_usage_bytes": [
          {
            "value": 123,
            "timestamp": "2023-11-07T05:31:56Z"
          }
        ],
        "ephemeral_storage": {
          "usage_bytes": [
            {
              "value": 123,
              "timestamp": "2023-11-07T05:31:56Z"
            }
          ],
          "utilization": [
            {
              "value": 123,
              "timestamp": "2023-11-07T05:31:56Z"
            }
          ]
        }
      }
    ]
  }
}

Authorizations

Authorization · string · header · required

Pass your Baseten API key. Clients automatically send Authorization: Bearer <key>. Direct callers can also use Authorization: Api-Key <key>; both schemes are accepted.

Path Parameters

deployment_id · string · required

Body

application/json

Time-range request for trainer deployment metrics.

end_epoch_millis · integer | null

Epoch millis to end fetching metrics.

start_epoch_millis · integer | null

Epoch millis to start fetching metrics.

step_seconds · integer | null

Resolution of the returned series, in seconds. When omitted, a step is derived from the time range so large windows return fewer points.

time_divisor_seconds · integer | null

Unit of time for request-volume metrics, in seconds (e.g. 60 for requests/minute). Defaults to per-second.

Response

200 - application/json

Response for POST /v1/loops/deployments/<id>/metrics.

deployment_id · string · required

The trainer deployment ID.

metrics · LoopsDeploymentMetricsV1 · object · required

Metrics for the deployment.

Show child attributes