Schedule pods to GPU-HPN virtual nodes based on attribute labels - Container Compute Service

When you use GPU-HPN resource reservations to run pods, you can schedule the pods to the specified virtual node by using the GPU attribute labels of the virtual node. This topic describes the attribute labels of GPU-HPN nodes and how to run pods on a specified GPU model.

Prerequisites

ACS is activated.
The cluster is associated with a GPU-HPN capacity reservation.

GPU attribute labels

GPU-HPN virtual nodes in ACS clusters have the following labels.

Label	Description
alibabacloud.com/gpu-model-series	The GPU model used by the virtual node.
alibabacloud.com/node-series	The type of reserved resource of the virtual node.
alibabacloud.com/hpn-zone	The zone of the high-performance network.

Note

To view the supported GPU models, submit a ticket.

Schedule pods to the virtual node with the specified attribute

You can use the node selector or node affinity rules of Kubernetes to control the GPU model to which pods are scheduled.

The following example uses the node selector to schedule pods to the specified GPU model.

Create a file named tensorflow-mnist.yaml and add the following content to the file.

apiVersion: batch/v1
kind: Job
metadata:
  name: tensorflow-mnist
spec:
  parallelism: 1
  template:
    metadata:
      labels:
        app: tensorflow-mnist
    spec:
      nodeSelector:
        alibabacloud.com/gpu-model-series: "gpu-example" # Schedule the pods of the application to the gpu-example virtual node. The model is for reference only. 
      containers:
      - name: tensorflow-mnist
        image: registry.cn-beijing.aliyuncs.com/acs/tensorflow-mnist-sample:v1.5
        command:
        - python
        - tensorflow-sample-code/tfjob/docker/mnist/main.py
        - --max_steps=1000
        - --data_dir=tensorflow-sample-code/data
        resources:
          requests:
            cpu: 1
            memory: 1
            nvidia.com/gpu: 1
          limits:
            cpu: 1
            memory: 1
            nvidia.com/gpu: 1
        workingDir: /root
      restartPolicy: Never

Run the following command to deploy the tensorflow-mnist application.
```
kubectl apply -f tensorflow-mnist.yaml
```

Run the following command to query the status of the pods.

kubectl get pod -l app=tensorflow-mnist -o wide

Expected results:

NAMESPACE   NAME                     READY   STATUS      RESTARTS   AGE    IP         NODE                                      NOMINATED NODE   READINESS GATES
default     tensorflow-mnist-xxx     0/2     Running     0          4h2m   <none>     cn-shanghai-b.cr-u4ub6c3un2mrjlct2l9c     <none>           <none>

The output indicates that the pods run on the virtual node that uses the gpu-example GPU.