deployments

Creates, updates, deletes, gets or lists a deployments resource.

Overview

Name	deployments
Type	Resource
Id	nvidia.nvcf_deployments.deployments

Fields

The following fields are returned by SELECT queries:

get

Name	Datatype	Description
deployment_id	string (uuid)	Last deployment id (wire: deploymentId)
function_id	string (uuid)	Function id (wire: functionId)
function_version_id	string (uuid)	Function version id (wire: functionVersionId)
nca_id	string (UTF-8)	NVIDIA Cloud Account Id (wire: ncaId)
function_name	string (UTF-8)	Function name (wire: functionName)
created_at	string (date_time)	Function deployment creation timestamp (wire: createdAt)
deployment_specifications	array	Function deployment details (wire: deploymentSpecifications)
function_status	string (UTF-8)	Function status (ACTIVE, DEPLOYING, ERROR, INACTIVE, DELETED, DEGRADED, DEGRADING) (wire: functionStatus)
health_info	array	Health info for a deployment specification is included only if there are any issues/errors. (wire: healthInfo)
last_updated_at	string (date_time)	Function deployment modification timestamp (wire: lastUpdatedAt)

get_by_id

Name	Datatype	Description
deployment_id	string (uuid)	Last deployment id (wire: deploymentId)
function_id	string (uuid)	Function id (wire: functionId)
function_version_id	string (uuid)	Function version id (wire: functionVersionId)
nca_id	string (UTF-8)	NVIDIA Cloud Account Id (wire: ncaId)
function_name	string (UTF-8)	Function name (wire: functionName)
created_at	string (date_time)	Function deployment creation timestamp (wire: createdAt)
deployment_specifications	array	Function deployment details (wire: deploymentSpecifications)
function_status	string (UTF-8)	Function status (ACTIVE, DEPLOYING, ERROR, INACTIVE, DELETED, DEGRADED, DEGRADING) (wire: functionStatus)
health_info	array	Health info for a deployment specification is included only if there are any issues/errors. (wire: healthInfo)
last_updated_at	string (date_time)	Function deployment modification timestamp (wire: lastUpdatedAt)

Methods

The following methods are available for this resource:

Name	Accessible by	Required Params	Optional Params	Description
get	select	function_id, function_version_id		Allows Account Admins to retrieve the deployment details of the specified function version. Access to this endpoint mandates a bearer token with 'deploy_function' scope in the HTTP Authorization header.
get_by_id	select	deployment_id		Allows Account Admins to retrieve the deployment details of the specified deployment id. Access to this endpoint mandates a bearer token with 'deploy_function' scope in the HTTP Authorization header.
create	insert	function_id, function_version_id, deployment_specifications		Initiates deployment for the specified function version. Upon invocation of this endpoint, the function's status transitions to 'DEPLOYING'. If the specified function version 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.
update	update	function_id, function_version_id, deployment_specifications		Deprecated and will be removed soon. Use a single GPU specification update method instead. Updates the deployment specs of the specified function version. 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.
delete	delete	function_id, function_version_id	graceful	Deletes the deployment associated with the specified function. Upon deletion, any active instances will be terminated, and the function's status will transition to 'INACTIVE'. To undeploy a function version gracefully, specify 'graceful=true' query parameter, allowing current tasks to complete before terminating the instances. If the specified function version 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.

Parameters

Parameters can be passed in the WHERE clause of a query. Check the Methods section to see which parameters are required or optional for each operation.

Name	Datatype	Description
deployment_id	string (uuid)	Deployment id
function_id	string (uuid)	Function id
function_version_id	string (uuid)	Function version id
graceful	boolean	Query param to deactivate function for graceful shutdown

SELECT examples

get

Allows Account Admins to retrieve the deployment details of the specified function version. Access to this endpoint mandates a bearer token with 'deploy_function' scope in the HTTP Authorization header.

SELECT
deployment_id,
function_id,
function_version_id,
nca_id,
function_name,
created_at,
deployment_specifications,
function_status,
health_info,
last_updated_at
FROM nvidia.nvcf_deployments.deployments
WHERE function_id = '{{ function_id }}' -- required
AND function_version_id = '{{ function_version_id }}' -- required
;

get_by_id

Allows Account Admins to retrieve the deployment details of the specified deployment id. Access to this endpoint mandates a bearer token with 'deploy_function' scope in the HTTP Authorization header.

SELECT
deployment_id,
function_id,
function_version_id,
nca_id,
function_name,
created_at,
deployment_specifications,
function_status,
health_info,
last_updated_at
FROM nvidia.nvcf_deployments.deployments
WHERE deployment_id = '{{ deployment_id }}' -- required
;

INSERT examples

create

Initiates deployment for the specified function version. Upon invocation of this endpoint, the function's status transitions to 'DEPLOYING'. If the specified function version 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.

INSERT INTO nvidia.nvcf_deployments.deployments (
deployment_specifications,
function_id,
function_version_id
)
SELECT 
'{{ deployment_specifications }}' /* required */,
'{{ function_id }}',
'{{ function_version_id }}'
RETURNING
deployment
;

Manifest

# Description fields are for documentation purposes
- name: deployments
props:
  - name: function_id
    value: "{{ function_id }}"
    description: Required parameter for the deployments resource.
  - name: function_version_id
    value: "{{ function_version_id }}"
    description: Required parameter for the deployments resource.
  - name: deployment_specifications
    description: |
      GPU specs with GPU, instance_type, clusters etc. details
    value:
      - attributes: "{{ attributes }}"
        autoscalingConfiguration:
          scaleDownDetails:
            factor: {{ factor }}
            metric: "{{ metric }}"
            stickiness:
              size: "{{ size }}"
              threshold: "{{ threshold }}"
            threshold: {{ threshold }}
          scaleUpDetails:
            factor: {{ factor }}
            metric: "{{ metric }}"
            stickiness:
              size: "{{ size }}"
              threshold: "{{ threshold }}"
            threshold: {{ threshold }}
        availabilityZones: "{{ availabilityZones }}"
        backend: "{{ backend }}"
        clusters: "{{ clusters }}"
        configuration: "{{ configuration }}"
        cpuArch: "{{ cpuArch }}"
        driverVersion: "{{ driverVersion }}"
        gpu: "{{ gpu }}"
        gpuMemory: "{{ gpuMemory }}"
        gpuSpecificationId: "{{ gpuSpecificationId }}"
        helmValidationPolicy:
          extraKubernetesTypes:
            - group: "{{ group }}"
              kind: "{{ kind }}"
              version: "{{ version }}"
          name: "{{ name }}"
        instanceType: "{{ instanceType }}"
        maxInstances: {{ maxInstances }}
        maxRequestConcurrency: {{ maxRequestConcurrency }}
        minInstances: {{ minInstances }}
        os: "{{ os }}"
        regions: "{{ regions }}"
        storage: "{{ storage }}"
        systemMemory: "{{ systemMemory }}"

UPDATE examples

update

Deprecated and will be removed soon. Use a single GPU specification update method instead. Updates the deployment specs of the specified function version. 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.

UPDATE nvidia.nvcf_deployments.deployments
SET 
deployment_specifications = '{{ deployment_specifications }}'
WHERE 
function_id = '{{ function_id }}' --required
AND function_version_id = '{{ function_version_id }}' --required
AND deployment_specifications = '{{ deployment_specifications }}' --required
RETURNING
deployment;

DELETE examples

delete

Deletes the deployment associated with the specified function. Upon deletion, any active instances will be terminated, and the function's status will transition to 'INACTIVE'. To undeploy a function version gracefully, specify 'graceful=true' query parameter, allowing current tasks to complete before terminating the instances. If the specified function version 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.

DELETE FROM nvidia.nvcf_deployments.deployments
WHERE function_id = '{{ function_id }}' --required
AND function_version_id = '{{ function_version_id }}' --required
AND graceful = '{{ graceful }}'
;