When this policy is enabled, applications in your ACK cluster can use native Kubernetes features like pod affinity, topology spreading, and node affinity for high availability and low latency.
Prerequisites
An ACK Pro cluster that runs Kubernetes 1.22 or later is required. For more information, see Create an ACK managed cluster and update a cluster.
Usage notes
-
Configuring pod anti-affinity can prevent a pod from being scheduled. Use this feature with caution.
-
Configuring virtual node anti-affinity can prevent a pod from being scheduled. Verify your configuration before you submit the pod.
-
When this feature is enabled, the scheduler queries inventory and updates the cluster status during pod scheduling. This process adds approximately 1 second of overhead for each pod. The maximum concurrent scheduling throughput is about 300 pods per second. If high scheduling speed or throughput is a priority, consider these performance impacts.
Step 1: Verify component installation and versions
This feature requires the Kube Scheduler and ACK Virtual Node components. Follow these steps to verify that they are installed and meet the version requirements.
Log on to the ACK console. In the left navigation pane, click Clusters.
On the Clusters page, click the name of your cluster. In the left navigation pane, click Add-ons.
-
In the Core Components section, verify that the following components are installed and meet the required versions. If not, click Install or Upgrade in the lower-right corner of the component's card.
-
Kube Scheduler: 5.9 or later
-
ACK Virtual Node: v2.10.0 or later
-
Step 2: Enable virtual node pod scheduling
-
Click Configuration in the lower-right corner of the Kube Scheduler component card.
-
Select the Enable Virtual Node-based Pod Scheduling checkbox and click OK.
