Container Compute Service:Use a custom scheduler

Step 1: Deploy a custom scheduler in the ACS cluster

Deploy a custom scheduler in the cluster. For more information, see the Kubernetes documentation.

Step 2: Modify the ACS scheduler configuration

1. Log on to the ACS console. In the left-side navigation pane, click Clusters.

2. On the Clusters page, find the target cluster and click its ID. In the left navigation pane, choose Operations > Add-ons.

3. On the Add-ons page, on the Kube Scheduler card, click Configuration.

4. In the dialog box that appears, select Enable custom labels and schedulers for GPU-HPN nodes, and then click OK.

   The following table describes the effects on pods and nodes when this option is enabled or disabled.

   Note

   This applies only to GPU-HPN pods and nodes. Other pod compute types do not support custom schedulers.

   	Using the default scheduler (option disabled)	Using a custom scheduler (option enabled)
   Pod scheduler name	Customization is not supported. After you submit a pod, the spec.schedulerName field is changed to default-scheduler.	Customization is supported. You can set the spec.schedulerName field of the pod to any value. The field is not modified after submission.
   Pod scheduling process	The ACS default scheduler allocates resources for all pods.	The ACS default scheduler only schedules pods where spec.schedulerName is set to default-scheduler. Custom schedulers allocate resources for all other pods.
   GPU-HPN node label and taint constraints	Adding, modifying, and deleting node labels, annotations, and taints are subject to ACS configuration constraints. For more information, see Manage node labels and taints.	Adding, modifying, and deleting node labels, annotations, and taints are no longer subject to ACS configuration constraints.
   Pod affinity scheduling constraints	The configuration of affinity fields is subject to ACS configuration constraints. For more information, see Node affinity scheduling.	The configuration of affinity fields is no longer subject to ACS configuration constraints.

Step 3: Configure a custom scheduler for a pod

1. Create a file named dep-with-koordinator.yaml with the following content.

   The file describes a deployment. The pod template specifies a custom scheduler named koord-scheduler.

   apiVersion: apps/v1
   kind: Deployment
   metadata:
     name: dep-with-koordinator
     labels:
       app: dep-with-koordinator
   spec:
     replicas: 1
     selector:
       matchLabels:
         app: dep-with-koordinator
     template:
       metadata:
         labels:
           app: dep-with-koordinator
           # Set the compute class to gpu-hpn. Other compute types do not support custom schedulers.
           alibabacloud.com/compute-class: gpu-hpn
       spec:
         containers:
         - name: demo
           image: registry.cn-hangzhou-finance.aliyuncs.com/acs/stress:v1.0.4
           command:
           - "sleep"
           - "infinity"
         restartPolicy: Always
         # Set the scheduler name to koord-scheduler. The name must match the one configured in Step 1.
         schedulerName: koord-scheduler

2. Submit the pod to the ACS cluster.

   kubectl apply -f dep-with-koordinator.yaml

3. Check the configured schedulerName of the pod.

   kubectl get pod -lapp=dep-with-koordinator -o custom-columns=NAME:.metadata.name,schedulerName:.spec.schedulerName

   Expected output:

   # The pod specifies the custom scheduler named koord-scheduler.
   NAME                               schedulerName
   dep-with-koordinator-xxxxx-xxxxx   koord-scheduler