Updated on 2026-09-01 GMT+08:00

Querying the Real-Time Resource Usage

Function

This API is used to obtain the real-time usage of all resource pools in the current project. This API is used when you need to query the real-time usage of resource pools when monitoring resource usage, optimizing resources, planning capacity, or troubleshooting faults. The resource usage includes the usage of CPU, memory, and storage resources. Before using this API, ensure that you have the required permission and there is a running resource pool. After the query, the API returns the real-time usage data of the resource pool, including the usage percentage, resource type, and timestamp. If you do not have the required permission, the resource pool does not exist, or the system cannot obtain the real-time data, the API will return an error message.

Debugging

You can debug this API through automatic authentication in API Explorer or use the SDK sample code generated by API Explorer.

Authorization Information

Each account has all the permissions required to call all APIs, but IAM users must be assigned the required permissions.

  • If you are using role/policy-based authorization, see Permissions Policies and Supported Actions for details on the required permissions.
  • If you are using identity policy-based authorization, the following identity policy-based permissions are required.

    Action

    Access Level

    Resource Type (*: required)

    Condition Key

    Alias

    Dependencies

    modelarts:pool:list

    List

    pool *

    -

    -

    -

    -

    • g:EnterpriseProjectId

    • g:RequestTag/<tag-key>

    • g:TagKeys

URI

GET /v2/{project_id}/metrics/runtime/pools

Table 1 Path Parameters

Parameter

Mandatory

Type

Description

project_id

Yes

String

Definition: User project ID. For details, see Obtaining a Project ID and Name.

Constraints: N/A

Range: N/A

Default Value: N/A

Request Parameters

None

Response Parameters

Status code: 200

Table 2 Response body parameters

Parameter

Type

Description

apiVersion

String

Definition: Resource version.

Range:

  • os.modelarts.xxxxx/v2

kind

String

Definition: Resource type.

Range:

  • PoolMetricsList

items

Array of MetricsItem objects

Definition: Metric list.

Table 3 MetricsItem

Parameter

Type

Description

table

MetricTableItem object

Definition: Resource list.

metadata

ResourceMetricsMetadata object

Definition: Resource metric metadata.

Table 4 MetricTableItem

Parameter

Type

Description

allocated

Allocated object

Definition: Allocated resources.

capacity

Capacity object

Definition: Resource capacity.

available

Available object

Definition: Available resource capacity, excluding the allocated resources, that is, the total resource capacity minus the amount of faulty resources and hot standby nodes.

workload

Workload object

Definition: Workload statistics.

Table 5 Allocated

Parameter

Type

Description

value

Value object

Definition: Number of resources.

timestamp

String

Definition: UTC time, in the format of yyyy-MM-dd'T'HH:mm:ss'Z'.

Range: N/A

window

String

Definition: Statistics interval. 1s indicates 1 second, 1m indicates 1 minute, and 1h indicates 1 hour.

Range: N/A

Table 6 Capacity

Parameter

Type

Description

value

Value object

Definition: Resource capacity minus the number of faulty resources.

maxValue

Value object

Definition: Maximum elastic resource capacity minus the number of faulty resources.

allValue

Value object

Definition: Number of resources.

allMaxValue

Value object

Definition: Maximum elastic resource capacity.

timestamp

String

Definition: UTC time, in the format of yyyy-MM-dd'T'HH:mm:ss'Z'.

Range: N/A

window

String

Definition: Statistics interval. 1s indicates 1 second, 1m indicates 1 minute, and 1h indicates 1 hour.

Range: N/A

Table 7 Available

Parameter

Type

Description

value

Value object

Resource capacity minus the amount of faulty resources and hot standby nodes.

maxValue

AnyType

Maximum elastic resource capacity minus the amount of faulty resources and hot standby nodes.

timestamp

String

UTC time, in the format of yyyy-MM-dd'T'HH:mm:ss'Z'.

window

String

Statistics interval. 1s indicates 1 second, 1m indicates 1 minute, and 1h indicates 1 hour.

Table 8 Value

Parameter

Type

Description

cpu

String

Definition: CPU quantity, that is, the amount of compute resources.

Range: N/A

memory

String

Definition: Memory.

Range: N/A

tnt004

String

Definition: Number of GPUs.

Range: N/A

Table 9 Workload

Parameter

Type

Description

allocated

WorkloadStatistics object

Workload statistics of delivered jobs.

queue

WorkloadStatistics object

Workload statistics of jobs in the queue.

timestamp

String

UTC time, in the format of yyyy-MM-dd'T'HH:mm:ss'Z'.

window

String

Statistics interval. 1s indicates 1 second, 1m indicates 1 minute, and 1h indicates 1 hour.

Table 10 WorkloadStatistics

Parameter

Type

Description

infer

Integer

Number of model deployment jobs of the old version.

notebook

Integer

Number of development environment jobs.

train

Integer

Number of training jobs.

warmUpTask

Integer

Number of weight warmup jobs.

x-infer

Integer

Number of model deployment jobs.

sum

Integer

Total number of jobs.

Table 11 ResourceMetricsMetadata

Parameter

Type

Description

name

String

Definition: Name of a resource metric.

Range: N/A

labels

Object

Definition: Tags of a resource metric.

Example Requests

GET https://{endpoint}/v2/{project_id}/metrics/runtime/pools

Example Responses

Status code: 200

OK.

{
  "apiVersion" : "v2",
  "kind" : "PoolMetricsList",
  "items" : [ {
    "table" : {
      "allocated" : {
        "value" : {
          "cpu" : "5",
          "memory" : "15548Mi",
          "tnt004" : "1073m"
        },
        "timestamp" : "2022-03-30T07:09:10Z",
        "window" : "1m"
      },
      "available" : {
        "value" : {
          "cpu" : 16,
          "memory" : "64Gi",
          "tnt004" : 2
        },
        "maxValue" : {
          "cpu" : 16,
          "memory" : "64Gi",
          "tnt004" : 2
        },
        "timestamp" : "2022-03-30T07:09:10Z",
        "window" : "1m"
      },
      "workload" : {
        "allocated" : {
          "infer" : 0,
          "train" : 2,
          "warmUpTask" : 1,
          "notebook" : 0,
          "x-infer" : 1,
          "sum" : 4
        },
        "queue" : {
          "infer" : 0,
          "train" : 3,
          "warmUpTask" : 0,
          "notebook" : 0,
          "x-infer" : 0,
          "sum" : 3
        },
        "timestamp" : "2022-03-30T07:09:10Z",
        "window" : "1m"
      },
      "capacity" : {
        "value" : {
          "cpu" : 16,
          "memory" : "64Gi",
          "tnt004" : 2
        },
        "maxValue" : {
          "cpu" : 16,
          "memory" : "64Gi",
          "tnt004" : 2
        },
        "allValue" : {
          "cpu" : 16,
          "memory" : "64Gi",
          "tnt004" : 2
        },
        "allMaxValue" : {
          "cpu" : 16,
          "memory" : "64Gi",
          "tnt004" : 2
        },
        "timestamp" : "2022-03-30T07:09:10Z",
        "window" : "1m"
      }
    },
    "metadata" : {
      "name" : "hougang-rse-pool"
    }
  } ]
}

Status Codes

Status Code

Description

200

OK.

Error Codes

See Error Codes.