All Products
Search
Document Center

Alibaba Cloud Linux:GetGpuClusterArgsOptions

Last Updated:Sep 14, 2026

Retrieves the GPU cluster parameter list.

Operation description

After installing the component for the target ACK cluster:

  1. First, when a cluster is managed for the first time, the component is installed on all ECS instances currently in the cluster. If the cluster has more than 50 nodes, only 50 instances are included in the first batch.

  2. Then, the operating system console periodically checks the scaling status of the managed cluster. When a new ECS instance is added to the cluster, the operating system console automatically installs the component without user intervention.

Try it now

Try this API in OpenAPI Explorer, no manual signing needed. Successful calls auto-generate SDK code matching your parameters. Download it with built-in credential security for local usage.

Test

RAM authorization

No authorization for this operation. If you encounter issues with this operation, contact technical support.

Request syntax

GET /api/v1/aiTopology/topology/listClusterTopologyOptions HTTP/1.1

Request parameters

Parameter

Type

Required

Description

Example

filterType

string

Yes

  1. Query the machine list of a cluster Set the retrieval type to comm_instance_id_list or instance_id_list. comm_instance_id_list: retrieves only the machines that participate in collective communication in the specified cluster. instance_id_list: retrieves all machines in the specified cluster.

  2. Query the AI task list of a cluster Set the retrieval type to comm_task_id_list or task_id_list. comm_task_id_list: retrieves only the tasks that involve collective communication in the specified cluster. task_id_list: retrieves all tasks in the specified cluster.

  3. Query the AI job pod list of a cluster Set the retrieval type to comm_pod_list or pod_list. comm_pod_list: retrieves only the pods that participate in collective communication in the specified cluster. pod_list: retrieves all pods in the specified cluster.

  4. Query the collective communication topology communication: retrieves all communication group IDs. channel: retrieves the channel list under the specified communication group ID. topoType: retrieves the topology type under the specified channel of the specified communication group ID.

instance_id_list

filterValues

any

No

The filter conditions. The value is a JSON object. The cluster_id field is required. The following optional filter conditions are supported:

  • instance_id_list

  • comm_instance_id_list

  • pod_list

  • comm_pod_list

  • communication

  • task_id_list

  • comm_task_id_list

  • channel

  • topoType

{ "cluster_id": "cluster1" }

Response elements

Element

Type

Description

Example

object

Schema of Response

code

string

The response code. Valid values: Success and Sysom.ServerError.

Success Fail

data

any

The result of the parameter retrieval.

["instance1", "instance2"]

message

string

The error message returned when the code is not Success.

some error msg while fail,“” if success

requestId

string

Id of the request

return real request id

total

integer

The total number of returned entries.

10

Examples

Success response

JSON format

{
  "code": "Success  Fail",
  "data": "[\"instance1\", \"instance2\"]",
  "message": "some error msg while fail,“” if success",
  "requestId": "return real request id",
  "total": 10
}

Error codes

HTTP status code

Error code

Error message

Description

400 Sysom.ServerError Server error, please correct the code according to the error message. %s. Server error, please correct the code according to the error message

See Error Codes for a complete list.

Release notes

See Release Notes for a complete list.