LWS

LWS#

LeaderWorkerSet (LWS) 是一个 Kubernetes API，旨在解决 AI/ML 推理工作负载的常见部署模式。一个主要的用例是多主机/多节点分布式推理。

vLLM 可以通过 Kubernetes 上的 LWS 部署，用于分布式模型服务。

准备工作#

至少需要两个 Kubernetes 节点，每个节点配备 8 个 GPU。
按照此处找到的说明安装 LWS。

部署和提供服务#

部署以下 yaml 文件 lws.yaml

apiVersion: leaderworkerset.x-k8s.io/v1
kind: LeaderWorkerSet
metadata:
  name: vllm
spec:
  replicas: 2
  leaderWorkerTemplate:
    size: 2
    restartPolicy: RecreateGroupOnPodRestart
    leaderTemplate:
      metadata:
        labels:
          role: leader
      spec:
        containers:
          - name: vllm-leader
            image: docker.io/vllm/vllm-openai:latest
            env:
              - name: HUGGING_FACE_HUB_TOKEN
                value: <your-hf-token>
            command:
              - sh
              - -c
              - "bash /vllm-workspace/examples/online_serving/multi-node-serving.sh leader --ray_cluster_size=$(LWS_GROUP_SIZE); 
                 python3 -m vllm.entrypoints.openai.api_server --port 8080 --model meta-llama/Meta-Llama-3.1-405B-Instruct --tensor-parallel-size 8 --pipeline_parallel_size 2"
            resources:
              limits:
                nvidia.com/gpu: "8"
                memory: 1124Gi
                ephemeral-storage: 800Gi
              requests:
                ephemeral-storage: 800Gi
                cpu: 125
            ports:
              - containerPort: 8080
            readinessProbe:
              tcpSocket:
                port: 8080
              initialDelaySeconds: 15
              periodSeconds: 10
            volumeMounts:
              - mountPath: /dev/shm
                name: dshm
        volumes:
        - name: dshm
          emptyDir:
            medium: Memory
            sizeLimit: 15Gi
    workerTemplate:
      spec:
        containers:
          - name: vllm-worker
            image: docker.io/vllm/vllm-openai:latest
            command:
              - sh
              - -c
              - "bash /vllm-workspace/examples/online_serving/multi-node-serving.sh worker --ray_address=$(LWS_LEADER_ADDRESS)"
            resources:
              limits:
                nvidia.com/gpu: "8"
                memory: 1124Gi
                ephemeral-storage: 800Gi
              requests:
                ephemeral-storage: 800Gi
                cpu: 125
            env:
              - name: HUGGING_FACE_HUB_TOKEN
                value: <your-hf-token>
            volumeMounts:
              - mountPath: /dev/shm
                name: dshm   
        volumes:
        - name: dshm
          emptyDir:
            medium: Memory
            sizeLimit: 15Gi
---
apiVersion: v1
kind: Service
metadata:
  name: vllm-leader
spec:
  ports:
    - name: http
      port: 8080
      protocol: TCP
      targetPort: 8080
  selector:
    leaderworkerset.sigs.k8s.io/name: vllm
    role: leader
  type: ClusterIP

kubectl apply -f lws.yaml

验证 Pod 的状态

kubectl get pods

应该得到类似于这样的输出

NAME       READY   STATUS    RESTARTS   AGE
vllm-0     1/1     Running   0          2s
vllm-0-1   1/1     Running   0          2s
vllm-1     1/1     Running   0          2s
vllm-1-1   1/1     Running   0          2s

验证分布式张量并行推理是否工作

kubectl logs vllm-0 |grep -i "Loading model weights took" 

应该得到类似于这样的结果

INFO 05-08 03:20:24 model_runner.py:173] Loading model weights took 0.1189 GB
(RayWorkerWrapper pid=169, ip=10.20.0.197) INFO 05-08 03:20:28 model_runner.py:173] Loading model weights took 0.1189 GB

访问 ClusterIP 服务#

# Listen on port 8080 locally, forwarding to the targetPort of the service's port 8080 in a pod selected by the service
kubectl port-forward svc/vllm-leader 8080:8080

输出应该类似于以下内容

Forwarding from 127.0.0.1:8080 -> 8080
Forwarding from [::1]:8080 -> 8080

提供模型服务#

打开另一个终端并发送请求

curl https://127.0.0.1:8080/v1/completions \
-H "Content-Type: application/json" \
-d '{
    "model": "meta-llama/Meta-Llama-3.1-405B-Instruct",
    "prompt": "San Francisco is a",
    "max_tokens": 7,
    "temperature": 0
}'