使用 GKE 推理网关提供 LLM

使用集合让一切井井有条 根据您的偏好保存内容并对其进行分类。

获取对模型的访问权限

如需将 Llama3.1 模型部署到 GKE，请签署许可同意协议并生成 Hugging Face 访问令牌。

签署许可同意协议

您必须签署同意协议才能使用 Llama3.1 模型。请按照以下说明操作：

1. 访问同意页面，使用您的 Hugging Face 账号验证同意情况。
2. 接受模型条款。

生成一个访问令牌

如需通过 Hugging Face 访问模型，您需要 Hugging Face 令牌。

如果您还没有令牌，请按照以下步骤生成新令牌：

1. 点击您的个人资料 > 设置 > 访问令牌。
2. 选择新建令牌 (New Token)。
3. 指定您选择的名称和一个至少为 Read 的角色。
4. 选择生成令牌。
5. 将生成的令牌复制到剪贴板。

准备环境

在本教程中，您将使用 Cloud Shell 来管理Google Cloud上托管的资源。Cloud Shell 预安装了本教程所需的软件，包括 kubectl 和 gcloud CLI。

如需使用 Cloud Shell 设置您的环境，请执行以下步骤：

1. 在 Google Cloud 控制台中，点击 Google Cloud 控制台中的 激活 Cloud Shell 以启动 Cloud Shell 会话。此操作会在 Google Cloud 控制台的底部窗格中启动会话。

2. 设置默认环境变量：

   gcloud config set project PROJECT_ID
   export PROJECT_ID=$(gcloud config get project)
   export REGION=REGION
   export CLUSTER_NAME=CLUSTER_NAME
   export HF_TOKEN=HF_TOKEN
   

   替换以下值：

   • PROJECT_ID：您的 Google Cloud项目 ID。
   • REGION：支持要使用的加速器类型的区域，例如适用于 H100 GPU 的 us-central1。
   • CLUSTER_NAME：您的集群的名称。
   • HF_TOKEN：您之前生成的 Hugging Face 令牌。

创建和配置 Google Cloud 资源

如需创建所需的资源，请按照以下说明操作。

创建 GKE 集群和节点池

在 GKE Autopilot 或 Standard 集群中的 GPU 上部署 LLM。我们建议您使用 Autopilot 集群获得全托管式 Kubernetes 体验。如需选择最适合您的工作负载的 GKE 操作模式，请参阅选择 GKE 操作模式。

gcloud container clusters create-auto CLUSTER_NAME \
    --project=PROJECT_ID \
    --region=REGION \
    --release-channel=rapid \
    --cluster-version=1.32.3-gke.1170000

1. 如需设置授权以抓取指标，请创建 inference-gateway-sa-metrics-reader-secret Secret：

   kubectl apply -f - <<EOF
   ---
   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRole
   metadata:
     name: inference-gateway-metrics-reader
   rules:
   - nonResourceURLs:
     - /metrics
     verbs:
     - get
   ---
   apiVersion: v1
   kind: ServiceAccount
   metadata:
     name: inference-gateway-sa-metrics-reader
     namespace: default
   ---
   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRoleBinding
   metadata:
     name: inference-gateway-sa-metrics-reader-role-binding
     namespace: default
   subjects:
   - kind: ServiceAccount
     name: inference-gateway-sa-metrics-reader
     namespace: default
   roleRef:
     kind: ClusterRole
     name: inference-gateway-metrics-reader
     apiGroup: rbac.authorization.k8s.io
   ---
   apiVersion: v1
   kind: Secret
   metadata:
     name: inference-gateway-sa-metrics-reader-secret
     namespace: default
     annotations:
       kubernetes.io/service-account.name: inference-gateway-sa-metrics-reader
   type: kubernetes.io/service-account-token
   ---
   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRole
   metadata:
     name: inference-gateway-sa-metrics-reader-secret-read
   rules:
   - resources:
     - secrets
     apiGroups: [""]
     verbs: ["get", "list", "watch"]
     resourceNames: ["inference-gateway-sa-metrics-reader-secret"]
   ---
   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRoleBinding
   metadata:
     name: gmp-system:collector:inference-gateway-sa-metrics-reader-secret-read
     namespace: default
   roleRef:
     name: inference-gateway-sa-metrics-reader-secret-read
     kind: ClusterRole
     apiGroup: rbac.authorization.k8s.io
   subjects:
   - name: collector
     namespace: gmp-system
     kind: ServiceAccount
   EOF
   

为 Hugging Face 凭据创建 Kubernetes Secret

在 Cloud Shell 中，执行以下操作：

1. 如需与集群通信，请配置 kubectl：

     gcloud container clusters get-credentials CLUSTER_NAME \
         --location=REGION
   

   替换以下值：

   • REGION：支持要使用的加速器类型的区域，例如适用于 H100 GPU 的 us-central1。
   • CLUSTER_NAME：您的集群的名称。

2. 创建包含 Hugging Face 令牌的 Kubernetes Secret：

     kubectl create secret generic HF_SECRET \
         --from-literal=token=HF_TOKEN \
         --dry-run=client -o yaml | kubectl apply -f -
   

   替换以下内容：

   • HF_TOKEN：您之前生成的 Hugging Face 令牌。
   • HF_SECRET：Kubernetes Secret 的名称。例如 hf-secret。

安装 InferenceModel 和 InferencePool CRD

在本部分中，您将为 GKE 推理网关安装必要的自定义资源定义 (CRD)。

CRD 会扩展 Kubernetes API。这样，您就可以定义新的资源类型。如需使用 GKE 推理网关，请运行以下命令，在 GKE 集群中安装 InferencePool 和 InferenceModel CRD：

kubectl apply -f https://github.com/kubernetes-sigs/gateway-api-inference-extension/releases/download/v0.3.0/manifests.yaml

部署模型服务器

此示例使用 vLLM 模型服务器部署 Llama3.1 模型。部署标记为 app:vllm-llama3-8b-instruct。此部署还使用了 Hugging Face 中名为 food-review 和 cad-fabricator 的两个 LoRA 适配器。您可以使用自己的模型服务器和模型容器、服务端口和部署名称更新此部署。您可以选择在部署中配置 LoRA 适配器，也可以部署基准模型。

1. 如需在 nvidia-h100-80gb 加速器类型上部署，请将以下清单保存为 vllm-llama3-8b-instruct.yaml。此清单定义了一个包含模型和模型服务器的 Kubernetes Deployment：

   apiVersion: apps/v1
   kind: Deployment
   metadata:
     name: vllm-llama3-8b-instruct
   spec:
     replicas: 3
     selector:
       matchLabels:
         app: vllm-llama3-8b-instruct
     template:
       metadata:
         labels:
           app: vllm-llama3-8b-instruct
       spec:
         containers:
           - name: vllm
             image: "vllm/vllm-openai:latest"
             imagePullPolicy: Always
             command: ["python3", "-m", "vllm.entrypoints.openai.api_server"]
             args:
             - "--model"
             - "meta-llama/Llama-3.1-8B-Instruct"
             - "--tensor-parallel-size"
             - "1"
             - "--port"
             - "8000"
             - "--enable-lora"
             - "--max-loras"
             - "2"
             - "--max-cpu-loras"
             - "12"
             env:
               # Enabling LoRA support temporarily disables automatic v1, we want to force it on
               # until 0.8.3 vLLM is released.
               - name: VLLM_USE_V1
                 value: "1"
               - name: PORT
                 value: "8000"
               - name: HUGGING_FACE_HUB_TOKEN
                 valueFrom:
                   secretKeyRef:
                     name: hf-token
                     key: token
               - name: VLLM_ALLOW_RUNTIME_LORA_UPDATING
                 value: "true"
             ports:
               - containerPort: 8000
                 name: http
                 protocol: TCP
             lifecycle:
               preStop:
                 # vLLM stops accepting connections when it receives SIGTERM, so we need to sleep
                 # to give upstream gateways a chance to take us out of rotation. The time we wait
                 # is dependent on the time it takes for all upstreams to completely remove us from
                 # rotation. Older or simpler load balancers might take upwards of 30s, but we expect
                 # our deployment to run behind a modern gateway like Envoy which is designed to
                 # probe for readiness aggressively.
                 sleep:
                   # Upstream gateway probers for health should be set on a low period, such as 5s,
                   # and the shorter we can tighten that bound the faster that we release
                   # accelerators during controlled shutdowns. However, we should expect variance,
                   # as load balancers may have internal delays, and we don't want to drop requests
                   # normally, so we're often aiming to set this value to a p99 propagation latency
                   # of readiness -> load balancer taking backend out of rotation, not the average.
                   #
                   # This value is generally stable and must often be experimentally determined on
                   # for a given load balancer and health check period. We set the value here to
                   # the highest value we observe on a supported load balancer, and we recommend
                   # tuning this value down and verifying no requests are dropped.
                   #
                   # If this value is updated, be sure to update terminationGracePeriodSeconds.
                   #
                   seconds: 30
                 #
                 # IMPORTANT: preStop.sleep is beta as of Kubernetes 1.30 - for older versions
                 # replace with this exec action.
                 #exec:
                 #  command:
                 #  - /usr/bin/sleep
                 #  - 30
             livenessProbe:
               httpGet:
                 path: /health
                 port: http
                 scheme: HTTP
               # vLLM's health check is simple, so we can more aggressively probe it.  Liveness
               # check endpoints should always be suitable for aggressive probing.
               periodSeconds: 1
               successThreshold: 1
               # vLLM has a very simple health implementation, which means that any failure is
               # likely significant. However, any liveness triggered restart requires the very
               # large core model to be reloaded, and so we should bias towards ensuring the
               # server is definitely unhealthy vs immediately restarting. Use 5 attempts as
               # evidence of a serious problem.
               failureThreshold: 5
               timeoutSeconds: 1
             readinessProbe:
               httpGet:
                 path: /health
                 port: http
                 scheme: HTTP
               # vLLM's health check is simple, so we can more aggressively probe it.  Readiness
               # check endpoints should always be suitable for aggressive probing, but may be
               # slightly more expensive than readiness probes.
               periodSeconds: 1
               successThreshold: 1
               # vLLM has a very simple health implementation, which means that any failure is
               # likely significant,
               failureThreshold: 1
               timeoutSeconds: 1
             # We set a startup probe so that we don't begin directing traffic or checking
             # liveness to this instance until the model is loaded.
             startupProbe:
               # Failure threshold is when we believe startup will not happen at all, and is set
               # to the maximum possible time we believe loading a model will take. In our
               # default configuration we are downloading a model from HuggingFace, which may
               # take a long time, then the model must load into the accelerator. We choose
               # 10 minutes as a reasonable maximum startup time before giving up and attempting
               # to restart the pod.
               #
               # IMPORTANT: If the core model takes more than 10 minutes to load, pods will crash
               # loop forever. Be sure to set this appropriately.
               failureThreshold: 3600
               # Set delay to start low so that if the base model changes to something smaller
               # or an optimization is deployed, we don't wait unnecessarily.
               initialDelaySeconds: 2
               # As a startup probe, this stops running and so we can more aggressively probe
               # even a moderately complex startup - this is a very important workload.
               periodSeconds: 1
               httpGet:
                 # vLLM does not start the OpenAI server (and hence make /health available)
                 # until models are loaded. This may not be true for all model servers.
                 path: /health
                 port: http
                 scheme: HTTP
   
             resources:
               limits:
                 nvidia.com/gpu: 1
               requests:
                 nvidia.com/gpu: 1
             volumeMounts:
               - mountPath: /data
                 name: data
               - mountPath: /dev/shm
                 name: shm
               - name: adapters
                 mountPath: "/adapters"
         initContainers:
           - name: lora-adapter-syncer
             tty: true
             stdin: true
             image: us-central1-docker.pkg.dev/k8s-staging-images/gateway-api-inference-extension/lora-syncer:main
             restartPolicy: Always
             imagePullPolicy: Always
             env:
               - name: DYNAMIC_LORA_ROLLOUT_CONFIG
                 value: "/config/configmap.yaml"
             volumeMounts: # DO NOT USE subPath, dynamic configmap updates don't work on subPaths
             - name: config-volume
               mountPath:  /config
         restartPolicy: Always
   
         # vLLM allows VLLM_PORT to be specified as an environment variable, but a user might
         # create a 'vllm' service in their namespace. That auto-injects VLLM_PORT in docker
         # compatible form as `tcp://<IP>:<PORT>` instead of the numeric value vLLM accepts
         # causing CrashLoopBackoff. Set service environment injection off by default.
         enableServiceLinks: false
   
         # Generally, the termination grace period needs to last longer than the slowest request
         # we expect to serve plus any extra time spent waiting for load balancers to take the
         # model server out of rotation.
         #
         # An easy starting point is the p99 or max request latency measured for your workload,
         # although LLM request latencies vary significantly if clients send longer inputs or
         # trigger longer outputs. Since steady state p99 will be higher than the latency
         # to drain a server, you may wish to slightly this value either experimentally or
         # via the calculation below.
         #
         # For most models you can derive an upper bound for the maximum drain latency as
         # follows:
         #
         #   1. Identify the maximum context length the model was trained on, or the maximum
         #      allowed length of output tokens configured on vLLM (llama2-7b was trained to
         #      4k context length, while llama3-8b was trained to 128k).
         #   2. Output tokens are the more compute intensive to calculate and the accelerator
         #      will have a maximum concurrency (batch size) - the time per output token at
         #      maximum batch with no prompt tokens being processed is the slowest an output
         #      token can be generated (for this model it would be about 100ms TPOT at a max
         #      batch size around 50)
         #   3. Calculate the worst case request duration if a request starts immediately
         #      before the server stops accepting new connections - generally when it receives
         #      SIGTERM (for this model that is about 4096 / 10 ~ 40s)
         #   4. If there are any requests generating prompt tokens that will delay when those
         #      output tokens start, and prompt token generation is roughly 6x faster than
         #      compute-bound output token generation, so add 20% to the time from above (40s +
         #      16s ~ 55s)
         #
         # Thus we think it will take us at worst about 55s to complete the longest possible
         # request the model is likely to receive at maximum concurrency (highest latency)
         # once requests stop being sent.
         #
         # NOTE: This number will be lower than steady state p99 latency since we stop       receiving
         #       new requests which require continuous prompt token computation.
             # NOTE: The max timeout for backend connections from gateway to model servers should
         #       be configured based on steady state p99 latency, not drain p99 latency
         #
         #   5. Add the time the pod takes in its preStop hook to allow the load balancers have
         #      stopped sending us new requests (55s + 30s ~ 85s)
         #
         # Because the termination grace period controls when the Kubelet forcibly terminates a
         # stuck or hung process (a possibility due to a GPU crash), there is operational safety
         # in keeping the value roughly proportional to the time to finish serving. There is also
         # value in adding a bit of extra time to deal with unexpectedly long workloads.
         #
         #   6. Add a 50% safety buffer to this time since the operational impact should be low
         #      (85s * 1.5 ~ 130s)
         #
         # One additional source of drain latency is that some workloads may run close to
         # saturation and have queued requests on each server. Since traffic in excess of the
         # max sustainable QPS will result in timeouts as the queues grow, we assume that failure
         # to drain in time due to excess queues at the time of shutdown is an expected failure
         # mode of server overload. If your workload occasionally experiences high queue depths
         # due to periodic traffic, consider increasing the safety margin above to account for
         # time to drain queued requests.
         terminationGracePeriodSeconds: 130
         nodeSelector:
           cloud.google.com/gke-accelerator: "nvidia-h100-80gb"
         volumes:
           - name: data
             emptyDir: {}
           - name: shm
             emptyDir:
               medium: Memory
           - name: adapters
             emptyDir: {}
           - name: config-volume
             configMap:
               name: vllm-llama3-8b-adapters
   ---
   apiVersion: v1
   kind: ConfigMap
   metadata:
     name: vllm-llama3-8b-adapters
   data:
     configmap.yaml: |
         vLLMLoRAConfig:
           name: vllm-llama3.1-8b-instruct
           port: 8000
           defaultBaseModel: meta-llama/Llama-3.1-8B-Instruct
           ensureExist:
             models:
             - id: food-review
               source: Kawon/llama3.1-food-finetune_v14_r8
             - id: cad-fabricator
               source: redcathode/fabricator
   ---
   kind: HealthCheckPolicy
   apiVersion: networking.gke.io/v1
   metadata:
     name: health-check-policy
     namespace: default
   spec:
     targetRef:
       group: "inference.networking.x-k8s.io"
       kind: InferencePool
       name: vllm-llama3-8b-instruct
     default:
       config:
         type: HTTP
         httpHealthCheck:
             requestPath: /health
             port: 8000
   

2. 将清单应用到您的集群：

   kubectl apply -f vllm-llama3-8b-instruct.yaml
   

创建 InferencePool 资源

InferencePool Kubernetes 自定义资源用于定义一组具有共同基础 LLM 和计算配置的 Pod。

InferencePool 自定义资源包含以下关键字段：

• selector：指定哪些 Pod 属于此池。此选择器中的标签必须与应用于模型服务器 Pod 的标签完全匹配。
• targetPort：定义 Pod 中模型服务器使用的端口。

InferencePool 资源可让 GKE 推理网关将流量路由到您的模型服务器 Pod。

如需使用 Helm 创建 InferencePool，请执行以下步骤：

helm install vllm-llama3-8b-instruct \
  --set inferencePool.modelServers.matchLabels.app=vllm-llama3-8b-instruct \
  --set provider.name=gke \
  --version v0.3.0 \
  oci://registry.k8s.io/gateway-api-inference-extension/charts/inferencepool

将以下字段更改为与您的部署相符：

• inferencePool.modelServers.matchLabels.app：用于选择模型服务器 Pod 的标签的键。

此命令会创建一个 InferencePool 对象，该对象在逻辑上表示模型服务器部署，并引用 Selector 选择的 Pod 中的模型端点服务。

创建具有投放重要性的 InferenceModel 资源

InferenceModel Kubernetes 自定义资源定义了特定模型（包括 LoRA 调优型模型）及其服务重要性。

InferenceModel 自定义资源包含以下关键字段：

• modelName：指定基准模型或 LoRA 适配器的名称。
• Criticality：指定模型的服务重要性。
• poolRef：引用模型的服务端 InferencePool。

借助 InferenceModel，GKE 推理网关可以根据模型名称和重要性将流量路由到您的模型服务器 pod。

如需创建 InferenceModel，请执行以下步骤：

1. 将以下示例清单保存为 inferencemodel.yaml：

   apiVersion: inference.networking.x-k8s.io/v1alpha2
   kind: InferenceModel
   metadata:
     name: inferencemodel-sample
   spec:
     modelName: MODEL_NAME
     criticality: CRITICALITY
     poolRef:
       name: INFERENCE_POOL_NAME
   

   替换以下内容：

   • MODEL_NAME：基准模型或 LoRA 适配器的名称。例如 food-review。
   • CRITICALITY：所选的广告投放重要性。从 Critical、Standard 或 Sheddable 中选择。例如 Standard。
   • INFERENCE_POOL_NAME：您在上一步中创建的 InferencePool 的名称。例如 vllm-llama3-8b-instruct。

2. 将示例清单应用到您的集群：

   kubectl apply -f inferencemodel.yaml
   

以下示例创建了一个 InferenceModel 对象，用于在 vllm-llama3-8b-instruct InferencePool 上配置 food-review LoRA 模型，并设置 Standard 服务重要性。InferenceModel 对象还会配置要以 Critical 优先级级别提供的基础模型。

apiVersion: inference.networking.x-k8s.io/v1alpha2
kind: InferenceModel
metadata:
  name: food-review
spec:
  modelName: food-review
  criticality: Standard
  poolRef:
    name: vllm-llama3-8b-instruct
  targetModels:
  - name: food-review
    weight: 100

---
apiVersion: inference.networking.x-k8s.io/v1alpha2
kind: InferenceModel
metadata:
  name: llama3-base-model
spec:
  modelName: meta-llama/Llama-3.1-8B-Instruct
  criticality: Critical
  poolRef:
    name: vllm-llama3-8b-instruct

创建网关

Gateway 资源充当外部流量进入 Kubernetes 集群的入口点。它定义了接受传入连接的监听器。

GKE 推理网关支持 gke-l7-rilb 和 gke-l7-regional-external-managed 网关类。如需了解详情，请参阅 GKE 文档中的网关类部分。

如需创建网关，请执行以下步骤：

1. 将以下示例清单保存为 gateway.yaml：

   apiVersion: gateway.networking.k8s.io/v1
   kind: Gateway
   metadata:
     name: GATEWAY_NAME
   spec:
     gatewayClassName: gke-l7-regional-external-managed
     listeners:
       - protocol: HTTP # Or HTTPS for production
         port: 80 # Or 443 for HTTPS
         name: http
   

   将 GATEWAY_NAME 替换为网关资源的唯一名称。例如 inference-gateway。

2. 将清单应用到您的集群：

   kubectl apply -f gateway.yaml
   

创建 HTTPRoute 资源

在本部分中，您将创建一个 HTTPRoute 资源，以定义网关如何将传入的 HTTP 请求路由到您的 InferencePool。

HTTPRoute 资源定义了 GKE 网关如何将传入 HTTP 请求路由到后端服务（即您的 InferencePool）。它用于指定匹配规则（例如标头或路径）以及应将流量转发到的后端。

如需创建 HTTPRoute，请执行以下步骤：

1. 将以下示例清单保存为 httproute.yaml：

   apiVersion: gateway.networking.k8s.io/v1
   kind: HTTPRoute
   metadata:
     name: HTTPROUTE_NAME
   spec:
     parentRefs:
     - name: GATEWAY_NAME
     rules:
     - matches:
       - path:
           type: PathPrefix
           value: PATH_PREFIX
       backendRefs:
       - name: INFERENCE_POOL_NAME
         group: inference.networking.x-k8s.io
         kind: InferencePool
   

   替换以下内容：

   • HTTPROUTE_NAME：您的 HTTPRoute 资源的唯一名称。例如 my-route。
   • GATEWAY_NAME：您创建的 Gateway 资源的名称。例如 inference-gateway。
   • PATH_PREFIX：用于匹配传入请求的路径前缀。例如，/ 用于匹配所有内容。
   • INFERENCE_POOL_NAME：您要将流量转送到的 InferencePool 资源的名称。例如 vllm-llama3-8b-instruct。

2. 将清单应用到您的集群：

   kubectl apply -f httproute.yaml
   

发送推理请求

配置 GKE 推理网关后，您可以向已部署的模型发送推理请求。

如需发送推理请求，请执行以下步骤：

• 检索网关端点。
• 构建格式正确的 JSON 请求。
• 使用 curl 将请求发送到 /v1/completions 端点。

这样，您就可以根据输入的提示和指定的参数生成文本。

1. 如需获取网关端点，请运行以下命令：

   IP=$(kubectl get gateway/GATEWAY_NAME -o jsonpath='{.status.addresses[0].address}')
   PORT=PORT_NUMBER # Use 443 for HTTPS, or 80 for HTTP
   

   替换以下内容：

   • GATEWAY_NAME：网关资源的名称。
   • PORT_NUMBER：您在网关中配置的端口号。

2. 如需使用 curl 向 /v1/completions 端点发送请求，请运行以下命令：

   curl -i -X POST https://${IP}:${PORT}/v1/completions \
   -H 'Content-Type: application/json' \
   -H 'Authorization: Bearer $(gcloud auth print-access-token)' \
   -d '{
       "model": "MODEL_NAME",
       "prompt": "PROMPT_TEXT",
       "max_tokens": MAX_TOKENS,
       "temperature": "TEMPERATURE"
   }'
   

   替换以下内容：

   • MODEL_NAME：要使用的模型或 LoRA 适配器的名称。
   • PROMPT_TEXT：模型的输入提示。
   • MAX_TOKENS：在回答中生成的词元数量上限。
   • TEMPERATURE：控制输出的随机性。使用�� 0 可获得确定性输出，使用更大的数字可获得更具创意的输出。

请注意以下行为：

• 请求正文：请求正文可以包含 stop 和 top_p 等其他参数。如需查看完整的选项列表，请参阅 OpenAI API 规范。
• 错误处理：在客户端代码中实现适当的错误处理，以处理响应中的潜在错误。例如，检查 curl 响应中的 HTTP 状态代码。非 200 状态代码通常表示存在错误。
• 身份验证和授权：对于生产部署，��使用身份验证和授权机制保护您的 API 端点。在请求中添加适当的标头（例如 Authorization）。

为推理网关配置可观测性

GKE 推理网关可让您了解推理工作负载的运行状况、性能和行为。这有助于您发现和解决问题、优化资源利用率，并确保应用的可靠性。您可以在 Cloud Monitoring 中通过 Metrics Explorer 查看这些可观测性指标。

如需为 GKE 推理网关配置可观测性，请参阅配置可观测性。

删除已部署的资源

为避免因您在本指南中创建的资源导致您的 Google Cloud 账号产生费用，请运行以下命令：

gcloud container clusters delete CLUSTER_NAME \
    --region=REGION

替换以下值：

• REGION：支持要使用的加速器类型的区域，例如适用于 H100 GPU 的 us-central1。
• CLUSTER_NAME：您的集群的名称。

后续步骤

• 了解 GKE 推理网关。
• 了解如何部署 GKE 推理网关。