使用 CLI v1 创建 Azure 机器学习计算群集

若要在 Python 中创建持久性 Azure 机器学习计算资源，请指定 vm_size 和 max_nodes 属性。 然后，Azure 机器学习将对其他属性使用智能默认值。

• vm_size：Azure 机器学习计算创建的节点的 VM 系列。
• max_nodes：在 Azure 机器学习计算中运行作业时自动扩展到的最大节点数。

适用于：Python SDK azureml v1

from azureml.core.compute import ComputeTarget, AmlCompute
from azureml.core.compute_target import ComputeTargetException

You can also configure several advanced properties when you create Azure Machine Learning Compute. The properties allow you to create a persistent cluster of fixed size, or within an existing Azure Virtual Network in your subscription.  See the [AmlCompute class](https://learn.microsoft.com/python/api/azureml-core/azureml.core.compute.amlcompute.amlcompute) for details.

> [!WARNING]
> When setting the `location` parameter, if it is a different region than your workspace or datastores you may see increased network latency and data transfer costs. The latency and costs can occur when creating the cluster, and when running jobs on it.

# [Azure CLI](#tab/azure-cli)

[!INCLUDE [cli v1](../includes/machine-learning-cli-v1.md)]

```azurecli
az ml computetarget create amlcompute -n cpu --min-nodes 1 --max-nodes 1 -s STANDARD_D3_V2 --location chinanorth2

有关详细信息，请参阅 Az PowerShell 模块 az ml computetarget create amlcompute。

适用于：Python SDK azureml v1

compute_config = AmlCompute.provisioning_configuration(vm_size='STANDARD_D2_V2',
                                                            vm_priority='lowpriority',
                                                            max_nodes=4)

适用于：Azure CLI ml 扩展 v1

设置 vm-priority：

az ml computetarget create amlcompute --name lowpriocluster --vm-size Standard_NC6 --max-nodes 5 --vm-priority lowpriority

适用于：Python SDK azureml v1

• 在预配配置中配置托管标识：

  • 在名为 ws 的工作区中创建的系统分配的托管标识

    # configure cluster with a system-assigned managed identity
    compute_config = AmlCompute.provisioning_configuration(vm_size='STANDARD_D2_V2',
                                                            max_nodes=5,
                                                            identity_type="SystemAssigned",
                                                            )
    cpu_cluster_name = "cpu-cluster"
    cpu_cluster = ComputeTarget.create(ws, cpu_cluster_name, compute_config)
    

  • 在名为 ws 的工作区中创建的用户分配的托管标识

    # configure cluster with a user-assigned managed identity
    compute_config = AmlCompute.provisioning_configuration(vm_size='STANDARD_D2_V2',
                                                            max_nodes=5,
                                                            identity_type="UserAssigned",
                                                            identity_id=['/subscriptions/<subcription_id>/resourcegroups/<resource_group>/providers/Microsoft.ManagedIdentity/userAssignedIdentities/<user_assigned_identity>'])
    
    cpu_cluster_name = "cpu-cluster"
    cpu_cluster = ComputeTarget.create(ws, cpu_cluster_name, compute_config)
    

• 将托管标识添加到名为 cpu_cluster 的现有计算群集

  • 系统分配的托管标识：

    # add a system-assigned managed identity
    cpu_cluster.add_identity(identity_type="SystemAssigned")
    

  • 用户分配的托管标识：

    # add a user-assigned managed identity
    cpu_cluster.add_identity(identity_type="UserAssigned", 
                                identity_id=['/subscriptions/<subcription_id>/resourcegroups/<resource_group>/providers/Microsoft.ManagedIdentity/userAssignedIdentities/<user_assigned_identity>'])
    

适用于：Azure CLI ml 扩展 v1

• 创建具有托管标识的新托管计算群集

  • 用户分配的托管标识

    az ml computetarget create amlcompute --name cpu-cluster --vm-size Standard_NC6 --max-nodes 5 --assign-identity '/subscriptions/<subcription_id>/resourcegroups/<resource_group>/providers/Microsoft.ManagedIdentity/userAssignedIdentities/<user_assigned_identity>'
    

  • 系统分配的托管标识

    az ml computetarget create amlcompute --name cpu-cluster --vm-size Standard_NC6 --max-nodes 5 --assign-identity '[system]'
    

• 将托管标识添加到现有群集：

  • 用户分配的托管标识

    az ml computetarget amlcompute identity assign --name cpu-cluster '/subscriptions/<subcription_id>/resourcegroups/<resource_group>/providers/Microsoft.ManagedIdentity/userAssignedIdentities/<user_assigned_identity>'
    

  • 系统分配的托管标识

    az ml computetarget amlcompute identity assign --name cpu-cluster '[system]'