Node Selection

It specifies a common algorithm for selecting nodes. Nodes are scored in simulated resource allocation to find the most suitable node for the current job.

Scoring parameters:

• nodeaffinity.weight: Pods are scheduled based on node affinity. Default value: 1
• podaffinity.weight: Pods are scheduled based on pod affinity. Default value: 1
• leastrequested.weight: Pods are scheduled to the node with the least requested resources. Default value: 1
• balancedresource.weight: Pods are scheduled to the node with balanced resource allocation. Default value: 1
• mostrequested.weight: Pods are scheduled to the node with the most requested resources. Default value: 0
• tainttoleration.weight: Pods are scheduled to the node with a high taint tolerance. Default value: 1
• imagelocality.weight: Pods are scheduled to the node where the required images exist. Default value: 1
• selectorspread.weight: Pods are evenly scheduled to different nodes. Default value: 0
• volumebinding.weight: Pods are scheduled to the node with the local PV delayed binding policy. Default value: 1
• podtopologyspread.weight: Pods are scheduled based on the pod topology. Default value: 2

Value Range

true: This plugin is enabled. false: This plugin is disabled.

Scoring parameters:

• nodeaffinity.weight: Enter a positive integer.
• podaffinity.weight: Enter a positive integer.
• leastrequested.weight: Enter a positive integer.
• balancedresource.weight: Enter a positive integer.
• mostrequested.weight: Enter a positive integer.
• tainttoleration.weight: Enter a positive integer.
• imagelocality.weight: Enter a positive integer.
• selectorspread.weight: Enter a positive integer.
• volumebinding.weight: Enter a positive integer.
• podtopologyspread.weight: Enter a positive integer.

Default Value

true

• nodeaffinity.weight: 1
• podaffinity.weight: 1
• leastrequested.weight: 1
• balancedresource.weight: 1
• mostrequested.weight: 0
• tainttoleration.weight: 1
• imagelocality.weight: 1
• selectorspread.weight: 0
• volumebinding.weight: 1
• podtopologyspread.weight: 2

Modifiable

Yes

Scope

CCE standard and CCE Turbo clusters

Suggestions

Use default settings. For specific requirements, adjust the weights based on your given service scenarios.

Increasing the value of a certain field will make it more influential during node scoring.

nodeorder and binpack magnify the basic scores of nodes based on the corresponding weights. If the weight of a field increases, the corresponding policy will have stronger influence on scoring a node. Other policies will be less influential.