Skip to content
DigitalOcean Gradient™ AI Agentic Cloud SDK Docs
  • Auto
  • Light
  • Dark

Autoscale

API Reference
GPU Droplets

Autoscale

Create a New Autoscale Pool
gpu_droplets.autoscale.create(AutoscaleCreateParams**kwargs) -> AutoscaleCreateResponse
post/v2/droplets/autoscale
Delete autoscale pool
gpu_droplets.autoscale.delete(strautoscale_pool_id)
delete/v2/droplets/autoscale/{autoscale_pool_id}
Delete autoscale pool and resources
gpu_droplets.autoscale.delete_dangerous(strautoscale_pool_id, AutoscaleDeleteDangerousParams**kwargs)
delete/v2/droplets/autoscale/{autoscale_pool_id}/dangerous
List All Autoscale Pools
gpu_droplets.autoscale.list(AutoscaleListParams**kwargs) -> AutoscaleListResponse
get/v2/droplets/autoscale
List history events
gpu_droplets.autoscale.list_history(strautoscale_pool_id, AutoscaleListHistoryParams**kwargs) -> AutoscaleListHistoryResponse
get/v2/droplets/autoscale/{autoscale_pool_id}/history
List members
gpu_droplets.autoscale.list_members(strautoscale_pool_id, AutoscaleListMembersParams**kwargs) -> AutoscaleListMembersResponse
get/v2/droplets/autoscale/{autoscale_pool_id}/members
Retrieve an Existing Autoscale Pool
gpu_droplets.autoscale.retrieve(strautoscale_pool_id) -> AutoscaleRetrieveResponse
get/v2/droplets/autoscale/{autoscale_pool_id}
Update Autoscale Pool
gpu_droplets.autoscale.update(strautoscale_pool_id, AutoscaleUpdateParams**kwargs) -> AutoscaleUpdateResponse
put/v2/droplets/autoscale/{autoscale_pool_id}
ModelsExpand Collapse
class AutoscalePool: …
id: str

A unique identifier for each autoscale pool instance. This is automatically generated upon autoscale pool creation.

active_resources_count: int

The number of active Droplets in the autoscale pool.

config: Config

The scaling configuration for an autoscale pool, which is how the pool scales up and down (either by resource utilization or static configuration).

created_at: datetime

A time value given in ISO8601 combined date and time format that represents when the autoscale pool was created.

formatdate-time
name: str

The human-readable name set for the autoscale pool.

status: Literal["active", "deleting", "error"]

The current status of the autoscale pool.

Accepts one of the following:
"active"
"deleting"
"error"
updated_at: datetime

A time value given in ISO8601 combined date and time format that represents when the autoscale pool was last updated.

formatdate-time
current_utilization: Optional[CurrentUtilization]
class AutoscalePoolDropletTemplate: …
image: str

The Droplet image to be used for all Droplets in the autoscale pool. You may specify the slug or the image ID.

region: Literal["nyc1", "nyc2", "nyc3", 11 more]

The datacenter in which all of the Droplets will be created.

Accepts one of the following:
"nyc1"
"nyc2"
"nyc3"
"ams2"
"ams3"
"sfo1"
"sfo2"
"sfo3"
"sgp1"
"lon1"
"fra1"
"tor1"
"blr1"
"syd1"
size: str

The Droplet size to be used for all Droplets in the autoscale pool.

ssh_keys: List[str]

The SSH keys to be installed on the Droplets in the autoscale pool. You can either specify the key ID or the fingerprint. Requires ssh_key:read scope.

ipv6: Optional[bool]

Assigns a unique IPv6 address to each of the Droplets in the autoscale pool.

name: Optional[str]

The name(s) to be applied to all Droplets in the autoscale pool.

project_id: Optional[str]

The project that the Droplets in the autoscale pool will belong to. Requires project:read scope.

tags: Optional[List[str]]

The tags to apply to each of the Droplets in the autoscale pool. Requires tag:read scope.

user_data: Optional[str]

A string containing user data that cloud-init consumes to configure a Droplet on first boot. User data is often a cloud-config file or Bash script. It must be plain text and may not exceed 64 KiB in size.

vpc_uuid: Optional[str]

The VPC where the Droplets in the autoscale pool will be created. The VPC must be in the region where you want to create the Droplets. Requires vpc:read scope.

with_droplet_agent: Optional[bool]

Installs the Droplet agent. This must be set to true to monitor Droplets for resource utilization scaling.

class AutoscalePoolDynamicConfig: …
max_instances: int

The maximum number of Droplets in an autoscale pool.

minimum1
maximum1000
min_instances: int

The minimum number of Droplets in an autoscale pool.

minimum1
maximum500
cooldown_minutes: Optional[int]

The number of minutes to wait between scaling events in an autoscale pool. Defaults to 10 minutes.

minimum5
maximum20
target_cpu_utilization: Optional[float]

Target CPU utilization as a decimal.

formatfloat
minimum0.05
maximum1
target_memory_utilization: Optional[float]

Target memory utilization as a decimal.

formatfloat
minimum0.05
maximum1
class AutoscalePoolStaticConfig: …
target_number_instances: int

Fixed number of instances in an autoscale pool.

minimum1
maximum1000
class CurrentUtilization: …
cpu: Optional[float]

The average CPU utilization of the autoscale pool.

formatfloat
minimum0
maximum1
memory: Optional[float]

The average memory utilization of the autoscale pool.

formatfloat
minimum0
maximum1