本文属于机器翻译版本。若本译文内容与英语原文存在差异，则一律以英文原文为准。

# 使用 Karpenter 自动扩缩功能创建和配置 HyperPod 集群
<a name="sagemaker-hyperpod-eks-autoscaling-cluster"></a>

在以下步骤中，您将创建一个启用了持续配置的 SageMaker HyperPod 集群，并将其配置为使用基于 Karpenter 的自动扩展。

**创建集 HyperPod 群**

1. 加载您的环境配置并从 CloudFormation堆栈中提取值。

   ```
   source .env
   SUBNET1=$(cfn-output $VPC_STACK_NAME PrivateSubnet1)
   SUBNET2=$(cfn-output $VPC_STACK_NAME PrivateSubnet2)
   SUBNET3=$(cfn-output $VPC_STACK_NAME PrivateSubnet3)
   SECURITY_GROUP=$(cfn-output $VPC_STACK_NAME NoIngressSecurityGroup)
   EKS_CLUSTER_ARN=$(cfn-output $EKS_STACK_NAME ClusterArn)
   EXECUTION_ROLE=$(cfn-output $SAGEMAKER_STACK_NAME ExecutionRole)
   SERVICE_ROLE=$(cfn-output $SAGEMAKER_STACK_NAME ServiceRole)
   BUCKET_NAME=$(cfn-output $SAGEMAKER_STACK_NAME Bucket)
   HP_CLUSTER_NAME="hyperpod-eks-test-$(date +%s)"
   EKS_CLUSTER_NAME=$(cfn-output $EKS_STACK_NAME ClusterName)
   HP_CLUSTER_ROLE=$(cfn-output $SAGEMAKER_STACK_NAME ClusterRole)
   ```

1. 将节点初始化脚本上传到 Amazon S3 存储桶。

   ```
   aws s3 cp lifecyclescripts/on_create_noop.sh s3://$BUCKET_NAME
   ```

1. 使用环境变量创建集群配置文件。

   ```
   cat > cluster_config.json << EOF
   {
       "ClusterName": "$HP_CLUSTER_NAME",
       "InstanceGroups": [
           {
               "InstanceCount": 1,
               "InstanceGroupName": "system",
               "InstanceType": "ml.c5.xlarge",
               "LifeCycleConfig": {
                   "SourceS3Uri": "s3://$BUCKET_NAME",
                   "OnCreate": "on_create_noop.sh"
               },
               "ExecutionRole": "$EXECUTION_ROLE"
           },
           {
               "InstanceCount": 0,
               "InstanceGroupName": "auto-c5-az1",
               "InstanceType": "ml.c5.xlarge",
               "LifeCycleConfig": {
                   "SourceS3Uri": "s3://$BUCKET_NAME",
                   "OnCreate": "on_create_noop.sh"
               },
               "ExecutionRole": "$EXECUTION_ROLE"
           },
           {
               "InstanceCount": 0,
               "InstanceGroupName": "auto-c5-4xaz2",
               "InstanceType": "ml.c5.4xlarge",
               "LifeCycleConfig": {
                   "SourceS3Uri": "s3://$BUCKET_NAME",
                   "OnCreate": "on_create_noop.sh"
               },
               "ExecutionRole": "$EXECUTION_ROLE",
               "OverrideVpcConfig": {
                   "SecurityGroupIds": [
                       "$SECURITY_GROUP"
                   ],
                   "Subnets": [
                       "$SUBNET2"
                   ]
               }
           },
           {
               "InstanceCount": 0,
               "InstanceGroupName": "auto-g5-az3",
               "InstanceType": "ml.g5.xlarge",
               "LifeCycleConfig": {
                   "SourceS3Uri": "s3://$BUCKET_NAME",
                   "OnCreate": "on_create_noop.sh"
               },
               "ExecutionRole": "$EXECUTION_ROLE",
               "OverrideVpcConfig": {
                   "SecurityGroupIds": [
                       "$SECURITY_GROUP"
                   ],
                   "Subnets": [
                       "$SUBNET3"
                   ]
               }
           }
       ],
       "VpcConfig": {
           "SecurityGroupIds": [
               "$SECURITY_GROUP"
           ],
           "Subnets": [
               "$SUBNET1"
           ]
       },
       "Orchestrator": {
           "Eks": {
               "ClusterArn": "$EKS_CLUSTER_ARN"
           }
       },
       "ClusterRole": "$HP_CLUSTER_ROLE",
       "AutoScaling": {
           "Mode": "Enable",
           "AutoScalerType": "Karpenter"
       },
       "NodeProvisioningMode": "Continuous"
   }
   EOF
   ```

1. 运行以下命令来创建您的 HyperPod 集群。

   ```
   aws sagemaker create-cluster --cli-input-json file://./cluster_config.json
   ```

1. 集群创建过程大约需要 20 分钟。监控集群状态，直到两者都 ClusterStatus 显示， AutoScaling.Status 都显示出来 InService。

1. 保存集群 ARN 以供后续操作使用。

   ```
   HP_CLUSTER_ARN=$(aws sagemaker describe-cluster --cluster-name $HP_CLUSTER_NAME \
      --output text --query ClusterArn)
   ```

**启用 Karpenter 自动扩缩**

1. 运行以下命令，在任意具有持续节点预调配模式的预先存在的集群上，启用基于 Karpenter 的自动扩缩。

   ```
   aws sagemaker update-cluster \
       --cluster-name $HP_CLUSTER_NAME \
       --auto-scaling Mode=Enable,AutoScalerType=Karpenter \
       --cluster-role $HP_CLUSTER_ROLE
   ```

1. 确认已成功启用 Karpenter：

   ```
   aws sagemaker describe-cluster --cluster-name $HP_CLUSTER_NAME --query 'AutoScaling'
   ```

1. 预期输出：

   ```
   {
       "Mode": "Enable",
       "AutoScalerType": "Karpenter",
       "Status": "InService"
   }
   ```

等待显示`Status``InService`后再继续配置 NodeClass 和 NodePool。