coscheduler

Documentation

Index

• Constants
• func NewCoscheduler(_ *runtime.Unknown, handle framework.FrameworkHandle) (framework.Plugin, error)
• type Coscheduler
  • func (c *Coscheduler) Filter(ctx context.Context, cycleState *framework.CycleState, pod *v1.Pod, ...) *framework.Status
  • func (c *Coscheduler) Less(podInfo1, podInfo2 *framework.PodInfo) bool
  • func (c *Coscheduler) Name() string
  • func (c *Coscheduler) PostBind(ctx context.Context, state *framework.CycleState, pod *v1.Pod, nodeName string)
  • func (c *Coscheduler) PreFilter(ctx context.Context, cycleState *framework.CycleState, pod *v1.Pod) *framework.Status
  • func (c *Coscheduler) PreFilterExtensions() framework.PreFilterExtensions
• type InsufficientResource
• type PodGroupInfo

Constants

const (
	// CoschedulerName is the name of the plugin used in the plugin registry and configurations.
	CoschedulerName = "CoschedulerSamenode"
	// PodGroupName is the name of a pod group that mark pods which should be scheduler to same node.
	PodGroupName = "pod-group.scheduling.sigs.k8s.io/name"
	// PodGroupTotal is the number of pods in the pod group
	PodGroupTotal = "pod-group.scheduling.sigs.k8s.io/total"
)

Functions

func NewCoscheduler

func NewCoscheduler(_ *runtime.Unknown, handle framework.FrameworkHandle) (framework.Plugin, error)

NewCoscheduler initializes a new plugin and returns it.

Types

type Coscheduler

type Coscheduler struct {
	// contains filtered or unexported fields
}

Coscheduler is a plugin that checks if a node has sufficient resources.

func (*Coscheduler) Filter

func (c *Coscheduler) Filter(ctx context.Context, cycleState *framework.CycleState, pod *v1.Pod, nodeInfo *schedulernodeinfo.NodeInfo) *framework.Status

Filter invoked at the filter extension point. Checks if a node has sufficient resources, such as cpu, memory, gpu, opaque int resources etc to run a pod. It returns a list of insufficient resources, if empty, then the node has all the resources requested by the pod.

func (*Coscheduler) Less

func (c *Coscheduler) Less(podInfo1, podInfo2 *framework.PodInfo) bool

Less is used to sort pods in the scheduling queue. 1. Compare the priorities of Pods. 2. Compare the initialization timestamps of PodGroups/Pods. 3. Compare the keys of PodGroups/Pods, i.e., if two pods are tied at priority and creation time, the one without podGroup will go ahead of the one with podGroup.

func (*Coscheduler) Name

func (c *Coscheduler) Name() string

Name returns name of the plugin. It is used in logs, etc.

func (*Coscheduler) PostBind

func (c *Coscheduler) PostBind(ctx context.Context, state *framework.CycleState, pod *v1.Pod, nodeName string)

PostBind is to clear Pginfo when every pod in the group is binded.

func (*Coscheduler) PreFilter

func (c *Coscheduler) PreFilter(ctx context.Context, cycleState *framework.CycleState, pod *v1.Pod) *framework.Status

PreFilter invoked at the prefilter extension point.

func (*Coscheduler) PreFilterExtensions

func (c *Coscheduler) PreFilterExtensions() framework.PreFilterExtensions

PreFilterExtensions returns prefilter extensions, pod add and remove.

type InsufficientResource

type InsufficientResource struct {
	ResourceName v1.ResourceName
	// We explicitly have a parameter for reason to avoid formatting a message on the fly
	// for common resources, which is expensive for cluster autoscaler simulations.
	Reason    string
	Requested int64
	Used      int64
	Capacity  int64
}

InsufficientResource describes what kind of resource limit is hit and caused the pod to not fit the node.

type PodGroupInfo

type PodGroupInfo struct {
	// contains filtered or unexported fields
}

PodGroupInfo is a wrapper to a PodGroup with additional information. A PodGroup's priority, timestamp are set according to the values of the PodGroup's first pod that is added to the scheduling queue.