简体中文版经机器翻译而成,仅供参考。如与英语版出现任何冲突,应以英语版为准。

为 Triton 推理服务器部署客户端(自动化部署)

提供者 kevin-hoke

要为 Triton 推理服务器部署客户端,请完成以下步骤:

  1. 打开 VI 编辑器,为 Triton 客户端创建部署,然后调用文件 Triton_client.YAML

    ---
    apiVersion: apps/v1
    kind: Deployment
    metadata:
      labels:
        app: triton-client
      name: triton-client
      namespace: triton
    spec:
      replicas: 1
      selector:
        matchLabels:
          app: triton-client
          version: v1
      template:
        metadata:
          labels:
            app: triton-client
            version: v1
        spec:
          containers:
          - image: nvcr.io/nvidia/tritonserver:20.07- v1- py3-clientsdk
            imagePullPolicy: IfNotPresent
            name: triton-client
            resources:
              limits:
                cpu: "2"
                memory: 4Gi
              requests:
                cpu: "2"
                memory: 4Gi
  2. 部署客户端。

    kubectl apply -f triton_client.yaml