Function Deployment

Update the specified GPU specification under the specified deployment

Updates the GPU specification of the specified function version's deployment. It's important to note that GPU type and backend configurations cannot be modified through this endpoint. If the specified function is public, then Account Admin cannot perform this operation. Access to this endpoint mandates a bearer token with 'deploy_function' scope in the HTTP Authorization header.

patch/v2/nvcf/deployments/{deploymentId}/gpu-specifications/{gpuSpecId}

Request

  • Base URL: https://api.nvcf.nvidia.com
  • URL: https://api.nvcf.nvidia.com/v2/nvcf/deployments/{deploymentId}/gpu-specifications/{gpuSpecId}
  • Auth: none declared

Path parameters

deploymentIdstring uuid required

Deployment id

gpuSpecIdstring uuid required

GPU Specification id

Request body

maxInstancesinteger nullable

Maximum number of instances for the deployment

minInstancesinteger nullable

Minimum number of instances for the deployment

autoscalingConfigurationPolicy'CUSTOM_CONFIGURATION' | 'PLATFORM_CONFIGURATION' UTF-8 nullable

Autoscaling configuration policy. CUSTOM_CONFIGURATION (default): use the provided autoscalingConfig. PLATFORM_CONFIGURATION: remove custom config and use platform defaults.

Example request

{
  "autoscalingConfiguration": {
    "scaleUpDetails": {
      "stickiness": {
        "size": "PT30M",
        "threshold": "PT5M"
      }
    },
    "scaleDownDetails": {
      "stickiness": {
        "size": "PT30M",
        "threshold": "PT5M"
      }
    }
  }
}

Response

OK

Example response

{
  "gpuSpecification": {
    "autoscalingConfiguration": {
      "scaleUpDetails": {
        "stickiness": {
          "size": "PT30M",
          "threshold": "PT5M"
        }
      },
      "scaleDownDetails": {
        "stickiness": {
          "size": "PT30M",
          "threshold": "PT5M"
        }
      }
    }
  }
}

Changes

No recorded changes to this operation across all 1 revision of this API.