kubernetesrm

package
v0.0.0-...-3511abf Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Nov 2, 2023 License: Apache-2.0 Imports: 63 Imported by: 0

Documentation

Index

Constants

View Source 
const ResourceTypeNvidia = "nvidia.com/gpu"

ResourceTypeNvidia describes the GPU resource type.

Variables

This section is empty.

Functions

This section is empty.

Types

type ChangePosition 

type ChangePosition struct {
	PodID cproto.ID
}

ChangePosition notifies the pods actor of a position change and to resubmit the specified pod.

type ChangePriority 

type ChangePriority struct {
	PodID cproto.ID
}

ChangePriority notifies the pods actor of a priority change and to resubmit the specified pod.

type KillTaskPod 

type KillTaskPod struct {
	PodID cproto.ID
}

KillTaskPod notifies the pods actor to kill a pod.

type PodsInfo 

type PodsInfo struct {
	NumAgents      int
	SlotsAvailable int
}

PodsInfo contains information for pods.

type PreemptTaskPod 

type PreemptTaskPod struct {
	PodName string
}

PreemptTaskPod notifies the pods actor to preempt a pod.

type ResourceManager 

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

ResourceManager is a resource manager that manages k8s resources.

func New 

func New(
	system *actor.System,
	db *db.PgDB,
	rmConfigs *config.ResourceConfig,
	taskContainerDefaults *model.TaskContainerDefaultsConfig,
	opts *aproto.MasterSetAgentOptions,
	cert *tls.Certificate,
) *ResourceManager

New returns a new ResourceManager, which communicates with and submits work to a Kubernetes apiserver.

func (*ResourceManager) Allocate 

func (k *ResourceManager) Allocate(msg sproto.AllocateRequest) (*sproto.ResourcesSubscription, error)

Allocate implements rm.ResourceManager.

func (ResourceManager) DeleteJob 

func (ResourceManager) DeleteJob(sproto.DeleteJob) (sproto.DeleteJobResponse, error)

DeleteJob implements rm.ResourceManager.

func (*ResourceManager) DisableAgent 

func (k *ResourceManager) DisableAgent(
	req *apiv1.DisableAgentRequest,
) (resp *apiv1.DisableAgentResponse, err error)

DisableAgent prevents scheduling on a node and has the option to kill running jobs.

func (ResourceManager) DisableSlot 

func (k ResourceManager) DisableSlot(
	req *apiv1.DisableSlotRequest,
) (resp *apiv1.DisableSlotResponse, err error)

DisableSlot implements 'det slot disable...' functionality.

func (*ResourceManager) EnableAgent 

func (k *ResourceManager) EnableAgent(
	req *apiv1.EnableAgentRequest,
) (resp *apiv1.EnableAgentResponse, err error)

EnableAgent allows scheduling on a node that has been disabled.

func (ResourceManager) EnableSlot 

func (k ResourceManager) EnableSlot(
	req *apiv1.EnableSlotRequest,
) (resp *apiv1.EnableSlotResponse, err error)

EnableSlot implements 'det slot enable...' functionality.

func (ResourceManager) ExternalPreemptionPending 

func (ResourceManager) ExternalPreemptionPending(sproto.PendingPreemption) error

ExternalPreemptionPending implements rm.ResourceManager.

func (ResourceManager) GetAgent 

func (ResourceManager) GetAgent(*apiv1.GetAgentRequest) (*apiv1.GetAgentResponse, error)

GetAgent implements rm.ResourceManager.

func (*ResourceManager) GetAgents 

func (k *ResourceManager) GetAgents(msg *apiv1.GetAgentsRequest) (*apiv1.GetAgentsResponse, error)

GetAgents implements rm.ResourceManager.

func (*ResourceManager) GetAllocationSummaries 

func (k *ResourceManager) GetAllocationSummaries(
	msg sproto.GetAllocationSummaries,
) (map[model.AllocationID]sproto.AllocationSummary, error)

GetAllocationSummaries implements rm.ResourceManager.

func (*ResourceManager) GetAllocationSummary 

func (k *ResourceManager) GetAllocationSummary(msg sproto.GetAllocationSummary) (*sproto.AllocationSummary, error)

GetAllocationSummary implements rm.ResourceManager.

func (*ResourceManager) GetDefaultAuxResourcePool 

func (k *ResourceManager) GetDefaultAuxResourcePool(
	sproto.GetDefaultAuxResourcePoolRequest,
) (sproto.GetDefaultAuxResourcePoolResponse, error)

GetDefaultAuxResourcePool implements rm.ResourceManager.

func (*ResourceManager) GetDefaultComputeResourcePool 

func (k *ResourceManager) GetDefaultComputeResourcePool(
	sproto.GetDefaultComputeResourcePoolRequest,
) (sproto.GetDefaultComputeResourcePoolResponse, error)

GetDefaultComputeResourcePool implements rm.ResourceManager.

func (ResourceManager) GetExternalJobs 

func (ResourceManager) GetExternalJobs(sproto.GetExternalJobs) ([]*jobv1.Job, error)

GetExternalJobs implements rm.ResourceManager.

func (*ResourceManager) GetJobQ 

func (k *ResourceManager) GetJobQ(msg sproto.GetJobQ) (map[model.JobID]*sproto.RMJobInfo, error)

GetJobQ implements rm.ResourceManager.

func (*ResourceManager) GetJobQueueStatsRequest 

func (k *ResourceManager) GetJobQueueStatsRequest(
	*apiv1.GetJobQueueStatsRequest,
) (*apiv1.GetJobQueueStatsResponse, error)

GetJobQueueStatsRequest implements rm.ResourceManager.

func (*ResourceManager) GetResourcePools 

func (k *ResourceManager) GetResourcePools(*apiv1.GetResourcePoolsRequest) (*apiv1.GetResourcePoolsResponse, error)

GetResourcePools implements rm.ResourceManager.

func (ResourceManager) GetSlot 

func (ResourceManager) GetSlot(*apiv1.GetSlotRequest) (*apiv1.GetSlotResponse, error)

GetSlot implements rm.ResourceManager. TODO(DET-9919): Implement GetSlot for Kubernetes RM.

func (ResourceManager) GetSlots 

func (ResourceManager) GetSlots(*apiv1.GetSlotsRequest) (*apiv1.GetSlotsResponse, error)

GetSlots implements rm.ResourceManager. TODO(DET-9919): Implement GetSlots for Kubernetes RM.

func (ResourceManager) IsReattachableOnlyAfterStarted 

func (k ResourceManager) IsReattachableOnlyAfterStarted() bool

IsReattachableOnlyAfterStarted always returns false for the k8s resource manager.

func (*ResourceManager) MoveJob 

func (k *ResourceManager) MoveJob(msg sproto.MoveJob) error

MoveJob implements rm.ResourceManager. TODO(DET-9920): This should know which pool it wants.

func (ResourceManager) NotifyContainerRunning 

func (k ResourceManager) NotifyContainerRunning(
	msg sproto.NotifyContainerRunning,
) error

NotifyContainerRunning receives a notification from the container to let the master know that the container is running.

func (*ResourceManager) RecoverJobPosition 

func (k *ResourceManager) RecoverJobPosition(msg sproto.RecoverJobPosition)

RecoverJobPosition implements rm.ResourceManager.

func (*ResourceManager) Release 

func (k *ResourceManager) Release(msg sproto.ResourcesReleased)

Release implements rm.ResourceManager. TODO(DET-9920): This should know which pool it wants.

func (ResourceManager) ResolveResourcePool 

func (k ResourceManager) ResolveResourcePool(
	name string,
	workspaceID int,
	slots int,
) (string, error)

ResolveResourcePool resolves the resource pool completely.

func (*ResourceManager) SetAllocationName 

func (k *ResourceManager) SetAllocationName(msg sproto.SetAllocationName)

SetAllocationName implements rm.ResourceManager. TODO(DET-9920): This should know which pool it wants.

func (*ResourceManager) SetGroupMaxSlots 

func (k *ResourceManager) SetGroupMaxSlots(msg sproto.SetGroupMaxSlots)

SetGroupMaxSlots implements rm.ResourceManager. TODO(DET-9920): This should know which pool it wants.

func (*ResourceManager) SetGroupPriority 

func (k *ResourceManager) SetGroupPriority(msg sproto.SetGroupPriority) error

SetGroupPriority implements rm.ResourceManager. TODO(DET-9920): This should know which pool it wants.

func (*ResourceManager) SetGroupWeight 

func (k *ResourceManager) SetGroupWeight(msg sproto.SetGroupWeight) error

SetGroupWeight implements rm.ResourceManager. TODO(DET-9920): This should know which pool it wants.

func (ResourceManager) TaskContainerDefaults 

func (k ResourceManager) TaskContainerDefaults(
	pool string,
	fallbackConfig model.TaskContainerDefaultsConfig,
) (result model.TaskContainerDefaultsConfig, err error)

TaskContainerDefaults returns TaskContainerDefaults for the specified pool.

func (*ResourceManager) ValidateCommandResources 

func (k *ResourceManager) ValidateCommandResources(
	msg sproto.ValidateCommandResourcesRequest,
) (sproto.ValidateCommandResourcesResponse, error)

ValidateCommandResources implements rm.ResourceManager.

func (ResourceManager) ValidateResourcePool 

func (k ResourceManager) ValidateResourcePool(name string) error

ValidateResourcePool validates that the named resource pool exists.

func (ResourceManager) ValidateResourcePoolAvailability 

func (k ResourceManager) ValidateResourcePoolAvailability(
	name string,
	slots int,
) ([]command.LaunchWarning, error)

ValidateResourcePoolAvailability checks the available resources for a given pool. This is a no-op for k8s.

func (ResourceManager) ValidateResources 

func (k ResourceManager) ValidateResources(
	name string,
	slots int,
	command bool,
) error

ValidateResources ensures enough resources are available in the resource pool. This is a no-op for k8s.

type StartTaskPod 

type StartTaskPod struct {
	Req          *sproto.AllocateRequest
	AllocationID model.AllocationID
	Spec         tasks.TaskSpec
	Slots        int
	Rank         int
	ResourcePool string
	Namespace    string

	LogContext logger.Context
}

StartTaskPod notifies the pods actor to start a pod with the task spec.

type SummarizeResources 

type SummarizeResources struct {
	PoolName string
}

SummarizeResources summerize pods resource.

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.