Use node instant scaling to automatically scale nodes and improve resource elasticity-Container Service for Kubernetes(ACK)-阿里云帮助中心

Before you begin

Review node scaling for these concepts:

How node instant scaling works
Benefits and supported use cases
Usage notes
To avoid unexpected charges, use pay-as-you-go instances. During scale-in, subscription instances are removed from the cluster but are not released from your account.

Prerequisites and limitations

An ACK managed cluster or ACK dedicated cluster running Kubernetes 1.24 or later. See Manually upgrade a cluster.
Auto Scaling is activated.
The vSwitch for the node instant scaling node pool must have sufficient IP addresses. Call DescribeVSwitchAttributes to check available IPs.

If IP addresses are insufficient, see Expand cluster IP capacity by adding a secondary CIDR block.
Node instant scaling works only with Standard Scaling Mode node pools. Swift mode is not supported.
With ACK GOATScaler v0.5.2 or earlier, manually remove offline nodes. See FAQ.
In an ACK dedicated cluster, nodes must have sufficient resources to deploy or update ACK GOATScaler; otherwise, scaling may fail.

Step 1: Enable node instant scaling

To use node instant scaling, enable cluster auto scaling on the Node Pools page.

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 Nodes > Node Pools.
On the Node Pools page, click Enable next to Node Scaling.
If prompted, activate Auto Scaling and grant the required permissions.
- ACK managed cluster: Authorize the AliyunCSManagedAutoScalerRole role.
- ACK dedicated cluster: Authorize the KubernetesWorkerRole role and attach the AliyunCSManagedAutoScalerRolePolicy.
  
  In the Node Scaling Configuration dialog box, after the precheck passes, click the RAM role link (such as KubernetesWorkerRole-xxxx) to complete authorization in the RAM console.

On the Node Scaling Configuration page, set Node Scaling Plan to Instant Scaling, configure the scaling parameters, and click OK.

The scaling component automatically triggers node scale-outs based on scheduling conditions.

Switch the Node Scaling Plan to Auto Scaling at any time. Follow the on-screen prompts. This feature is in beta. To participate, submit a ticket.

Parameter	Description
Scale-in Threshold	The ratio of resource requests to the total resource capacity of a node in a node pool with node autoscaling enabled. A node is eligible for scale-in only if its CPU and memory resource utilization are both below the Scale-in Threshold.
GPU Scale-in Threshold	The scale-in threshold for GPU instances. A GPU instance is eligible for scale-in only when its CPU, memory, and GPU utilization all fall below the configured GPU Scale-in Threshold.
Scale-in Trigger Delay	The delay between when a node becomes eligible for scale-in and when the scale-in operation is performed. Unit: minutes. Default value: 10 minutes. Important The scaling component can perform a node scale-in only after the Scale-in Threshold condition is met and the Scale-in Trigger Delay duration has passed.

Advanced configurations

Parameter	Description
Pod Termination Timeout	Maximum wait time for pod termination during scale-in. Unit: seconds. If a pod is not evicted before the timeout, the node is not released.
Minimum Pod Replicas	Scale-in protection threshold. Nodes with `ReplicationController` or `ReplicaSet` pods are not scaled in if the replica count falls below this value. Applies only to `ReplicationController` and `ReplicaSet` pods, not `StatefulSet` or `DaemonSet`.
Enable DaemonSet Pod Eviction	When enabled, DaemonSet pods are evicted when their node is scaled in.
Skip nodes with pods in the kube-system namespace	When enabled, nodes with kube-system namespace pods are excluded from scale-in. Note This does not apply to DaemonSet pods or mirror pods.

Step 2: Configure a node pool for auto scaling

Node instant scaling applies only to node pools with auto scaling enabled.

Create a node pool and set Scaling Mode to Auto.
Edit an existing node pool to change its Scaling Mode to Auto.

Recommended: Configure multiple instance types and availability zones to ensure sufficient capacity during scale-outs.

(Optional) Step 3: Verify the configuration

Verify that auto scaling is active and ACK GOATScaler is installed.

Verify node pool status

On the Node Pools page, ensure your node pool displays the Auto Scaling tag.

Verify add-on installation

On the Clusters page, click the name of your cluster. In the left navigation pane, click Components and Add-ons.
On the Add-ons page, ensure the ACK GOATScaler add-on status is Installed.

Node instant scaling key events

The node instant scaling feature generates these events:

Event name	Object	Description
ProvisionNode	`pod`	Node scale-out triggered successfully.
ProvisionNodeFailed	`pod`	Node scale-out failed to trigger.
ResetPod	`pod`	An unschedulable pod that triggered a scale-out is requeued for retry.
InstanceInventoryStatusChanged	`ACKNodePool`	Emitted when available inventory changes for configured instance types. See View the health status of node instant scaling.

Node instant scaling identifiers

These identifiers are system-managed. Modifying them may cause unexpected scaling behavior.

Node labels

goatscaler.io/managed:true or k8s.aliyun.com: true: Identifies nodes managed by node instant scaling. Used to evaluate scale-in conditions.
goatscaler.io/provision-task-id:{task-id}: The scale-out task ID. Used for tracing.

Node taints

goatscaler.io/node-terminating: Nodes with this taint are marked for scale-in.

Pod annotations

goatscaler.io/provision-task-id: The scale-out task ID for this pod. The system waits for the node to start before triggering more scale-outs.
goatscaler.io/reschedule-deadline: Pod scheduling timeout. If still unschedulable past this deadline, the pod is requeued and can trigger another scale-out.

FAQ

Category	Subcategory	Link
Node instant scaling behavior	Known limitations
	Scale-out behavior	What resource types does node instant scaling use for scaling simulations? Does node instant scaling select an appropriate instance type from a node pool based on pod resource requests? How does node instant scaling select an instance type from a node pool with multiple types? How do I monitor the inventory of instance types in a node pool when using node instant scaling? How can I optimize the node pool configuration to avoid scale-out failures from insufficient inventory? Why does node instant scaling fail to scale out nodes? How do I configure custom resources for a node pool with node instant scaling enabled?
	Scale-in behavior	Why does node instant scaling fail to scale in nodes? What types of pods can prevent node instant scaling from removing a node?
Custom scaling behavior	Controlling scaling with pods	How can I control node scale-in by using pods?
Custom scaling behavior	Controlling scaling with nodes	How do I specify which node to delete during a scale-in? How do I prevent a specific node from being scaled in? Can node instant scaling be configured to scale in only empty nodes?
Node instant scaling add-on		Does the node instant scaling add-on update automatically? Why does node scaling fail on my ACK managed cluster after I granted the required permissions?

Related operations

View health status

Thenode instant scalingfeature selects instance types and availability zones based on ECS inventory. Check the node pool ConfigMap to monitor inventory health and get instance type recommendations.

See View the health status of node instant scaling.

Enable log collection

For ACK managed clusters, collect ACK GOATScaler logs from the Control Plane Component Logs page.

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 Operations > Log Center.
On the Control Plane Component Logs tab, click Update Component and select ACK GOATScaler.

In the Update Component dialog box, select ack-goatscaler and click OK.

After the update, select ACK GOATScaler from the drop-down list to view logs.

Upgrade ACK GOATScaler

Keep ACK GOATScaler updated for the latest features. See Manage components.

Skip inventory checks for private pools

If you use a private pool for guaranteed capacity, enable SkipInventoryCheck to let ACK GOATScaler bypass inventory checks and use private pool resources directly.

On the Clusters page, click the name of your cluster. In the left navigation pane, click Components and Add-ons.
On the Core Component page, find ACK GOATScaler and click Configuration.

Requires ACK GOATScaler v0.3.0-582e405-aliyun or later.
Set SkipInventoryCheck to true.