devicemanager

type ActivePodsWithRanks ¶ added in v0.1.21

type ActivePodsWithRanks struct {
	Pods []*v1.Pod

	// Rank is a ranking of pods.  This ranking is used during sorting when
	// comparing two pods that are both scheduled, in the same phase, and
	// having the same ready status.
	Rank []int
	// Now is a reference timestamp for doing logarithmic timestamp comparisons.
	// If zero, comparison happens without scaling.
	Now metav1.Time
}

ActivePodsWithRanks is a sortable list of pods and a list of corresponding ranks which will be considered during sorting. The two lists must have equal length. After sorting, the pods will be ordered as follows, applying each rule in turn until one matches:

1. If only one of the pods is assigned to a node, the pod that is not assigned comes before the pod that is.
2. If the pods' phases differ, a pending pod comes before a pod whose phase is unknown, and a pod whose phase is unknown comes before a running pod.
3. If exactly one of the pods is ready, the pod that is not ready comes before the ready pod.
4. If controller.kubernetes.io/pod-deletion-cost annotation is set, then the pod with the lower value will come first.
5. If the pods' ranks differ, the pod with greater rank comes before the pod with lower rank.
6. If both pods are ready but have not been ready for the same amount of time, the pod that has been ready for a shorter amount of time comes before the pod that has been ready for longer.
7. If one pod has a container that has restarted more than any container in the other pod, the pod with the container with more restarts comes before the other pod.
8. If the pods' creation times differ, the pod that was created more recently comes before the older pod.

In 6 and 8, times are compared in a logarithmic scale. This allows a level of randomness among equivalent Pods when sorting. If two pods have the same logarithmic rank, they are sorted by UUID to provide a pseudorandom order.

If none of these rules matches, the second pod comes before the first pod.

The intention of this ordering is to put pods that should be preferred for deletion first in the list.

func (s ActivePodsWithRanks) Len() int

func (s ActivePodsWithRanks) Less(i, j int) bool

func (s ActivePodsWithRanks) Swap(i, j int)

type ClientStatus ¶

type ClientStatus string

const (
	ClientReady    ClientStatus = "Ready"
	ClientNotReady ClientStatus = "NotReady"
)

type Controller ¶

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

func NewController(
	kubeclient kubernetes.Interface,
	kubeshareclient clientset.Interface,
	nodeInformer coreinformers.NodeInformer,
	podInformer coreinformers.PodInformer,

	kubeshareInformer informers.SharePodInformer) *Controller

func (c *Controller) Run(threadiness int, stopCh <-chan struct{}) error

type GPUInfo ¶

type GPUInfo struct {
	UUID    string
	Usage   float64
	Mem     int64
	PodList *list.List
}

type GPUResource ¶ added in v0.1.21

type GPUResource struct {
	GPUFreeReq int64
	// GPUFreeMem in bytes
	GPUFreeMem int64

	GPUAffinityTags     []string
	GPUAntiAffinityTags []string
	// len(GPUExclusionTags) should be only one
	GPUExclusionTags []string
}

func (this *GPUResource) DeepCopy() *GPUResource

type NameList ¶ added in v0.1.22

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

TODO here to extend to new Struct for multil access for different sharepod

type NodeInfo ¶

type NodeInfo struct {
	// GPUID -> GPU
	GPUID2GPU map[string]*GPUInfo
	// UUID -> Port (string)
	UUID2Port map[string]string

	// port in use
	PodManagerPortBitmap *bitmap.RRBitmap
	PodIP                string
}

type NodeResource ¶ added in v0.1.21

type NodeResource struct {
	CpuTotal int64
	MemTotal int64
	GpuTotal int
	// GpuMemTotal in bytes
	GpuMemTotal int64
	CpuFree     int64
	MemFree     int64
	/* Available GPU calculate */
	// Total GPU count - Pods using nvidia.com/gpu
	GpuFreeCount int
	// GPUs available usage (1.0 - SharePod usage)
	// GPUID to integer index mapping
	GpuFree map[string]*GPUResource
}

func (this *NodeResource) DeepCopy() *NodeResource

type NodeResources ¶ added in v0.1.21

type NodeResources map[string]*NodeResource

NodeResources: Available resources in cluster to schedule Training Jobs

func (this *NodeResources) DeepCopy() *NodeResources

func (this *NodeResources) PrintMe()

type NodeStatus ¶

type NodeStatus struct {
	Conn          net.Conn
	LastHeartbeat time.Time
	ClientStatus  ClientStatus
}

type PodRequest ¶

type PodRequest struct {
	Key            string
	Request        float64
	Limit          float64
	Memory         int64
	PodManagerPort int
}

func FindInQueue(key string, pl *list.List) (*PodRequest, bool)

type RealPodControl ¶ added in v0.1.21

type RealPodControl struct {
	KubeClient clientset.Interface
	Recorder   record.EventRecorder
}

type SharepodProbes ¶ added in v0.1.23

type SharepodProbes struct {
	Liveness  *v1.Probe
	Readiness *v1.Probe
}