pytorch

Documentation

Index

• func ContainMasterSpec(job *v1.PyTorchJob) bool
• type PytorchJobReconciler
  • func NewReconciler(mgr ctrl.Manager, config job_controller.JobControllerConfiguration) *PytorchJobReconciler
  • func (r *PytorchJobReconciler) ControllerName() string
  • func (r *PytorchJobReconciler) CreatePod(job interface{}, pod *corev1.Pod) error
  • func (r *PytorchJobReconciler) CreateService(job interface{}, service *corev1.Service) error
  • func (r *PytorchJobReconciler) DeleteJob(job interface{}) error
  • func (r *PytorchJobReconciler) DeletePod(job interface{}, pod *corev1.Pod) error
  • func (r *PytorchJobReconciler) DeleteService(job interface{}, name string, namespace string) error
  • func (r *PytorchJobReconciler) GetAPIGroupVersion() schema.GroupVersion
  • func (r *PytorchJobReconciler) GetAPIGroupVersionKind() schema.GroupVersionKind
  • func (r *PytorchJobReconciler) GetDefaultContainerName() string
  • func (r *PytorchJobReconciler) GetDefaultContainerPortName() string
  • func (r *PytorchJobReconciler) GetDefaultContainerPortNumber() int32
  • func (r *PytorchJobReconciler) GetGroupNameLabelValue() string
  • func (r *PytorchJobReconciler) GetJobFromAPIClient(namespace, name string) (metav1.Object, error)
  • func (r *PytorchJobReconciler) GetJobFromInformerCache(namespace, name string) (metav1.Object, error)
  • func (r *PytorchJobReconciler) GetPodsForJob(obj interface{}) ([]*corev1.Pod, error)
  • func (r *PytorchJobReconciler) GetReconcileOrders() []v1.ReplicaType
  • func (r *PytorchJobReconciler) GetServicesForJob(obj interface{}) ([]*corev1.Service, error)
  • func (r *PytorchJobReconciler) IsMasterRole(replicas map[v1.ReplicaType]*v1.ReplicaSpec, rtype v1.ReplicaType, index int) bool
  • func (r *PytorchJobReconciler) Reconcile(req ctrl.Request) (ctrl.Result, error)
  • func (r *PytorchJobReconciler) SetClusterSpec(job interface{}, podTemplate *corev1.PodTemplateSpec, rtype, index string) error
  • func (r *PytorchJobReconciler) SetupWithManager(mgr ctrl.Manager) error
  • func (r *PytorchJobReconciler) UpdateJobStatus(job interface{}, replicas map[v1.ReplicaType]*v1.ReplicaSpec, ...) error
  • func (r *PytorchJobReconciler) UpdateJobStatusInApiServer(job interface{}, jobStatus *v1.JobStatus) error

Functions

func ContainMasterSpec

func ContainMasterSpec(job *v1.PyTorchJob) bool

Types

type PytorchJobReconciler

type PytorchJobReconciler struct {
	client.Client
	// contains filtered or unexported fields
}

PytorchJobReconciler reconcile a PytorchJob object

func NewReconciler

func NewReconciler(mgr ctrl.Manager, config job_controller.JobControllerConfiguration) *PytorchJobReconciler

func (*PytorchJobReconciler) ControllerName

func (r *PytorchJobReconciler) ControllerName() string

func (*PytorchJobReconciler) CreatePod

func (r *PytorchJobReconciler) CreatePod(job interface{}, pod *corev1.Pod) error

CreatePod creates the pod

func (*PytorchJobReconciler) CreateService

func (r *PytorchJobReconciler) CreateService(job interface{}, service *corev1.Service) error

CreateService creates the service

func (*PytorchJobReconciler) DeleteJob

func (r *PytorchJobReconciler) DeleteJob(job interface{}) error

DeleteJob deletes the job

func (*PytorchJobReconciler) DeletePod

func (r *PytorchJobReconciler) DeletePod(job interface{}, pod *corev1.Pod) error

DeletePod deletes the pod

func (*PytorchJobReconciler) DeleteService

func (r *PytorchJobReconciler) DeleteService(job interface{}, name string, namespace string) error

DeleteService deletes the service

func (*PytorchJobReconciler) GetAPIGroupVersion

func (r *PytorchJobReconciler) GetAPIGroupVersion() schema.GroupVersion

GetAPIGroupVersion returns the GroupVersion of the API

func (*PytorchJobReconciler) GetAPIGroupVersionKind

func (r *PytorchJobReconciler) GetAPIGroupVersionKind() schema.GroupVersionKind

GetAPIGroupVersionKind returns the GroupVersionKind of the API

func (*PytorchJobReconciler) GetDefaultContainerName

func (r *PytorchJobReconciler) GetDefaultContainerName() string

GetDefaultContainerName returns the default container name in pod

func (*PytorchJobReconciler) GetDefaultContainerPortName

func (r *PytorchJobReconciler) GetDefaultContainerPortName() string

GetDefaultContainerPortName Get the default container port name

func (*PytorchJobReconciler) GetDefaultContainerPortNumber

func (r *PytorchJobReconciler) GetDefaultContainerPortNumber() int32

GetDefaultContainerPortNumber get the default container port number

func (*PytorchJobReconciler) GetGroupNameLabelValue

func (r *PytorchJobReconciler) GetGroupNameLabelValue() string

GetGroupNameLabelValue returns the Group Name(value) in the labels of the job

func (*PytorchJobReconciler) GetJobFromAPIClient

func (r *PytorchJobReconciler) GetJobFromAPIClient(namespace, name string) (metav1.Object, error)

GetJobFromAPIClient returns the Job from API server

func (*PytorchJobReconciler) GetJobFromInformerCache

func (r *PytorchJobReconciler) GetJobFromInformerCache(namespace, name string) (metav1.Object, error)

GetJobFromInformerCache returns the Job from Informer Cache

func (*PytorchJobReconciler) GetPodsForJob

func (r *PytorchJobReconciler) GetPodsForJob(obj interface{}) ([]*corev1.Pod, error)

GetPodsForJob returns the pods managed by the job. This can be achieved by selecting pods using label key "job-name" i.e. all pods created by the job will come with label "job-name" = <this_job_name>

func (*PytorchJobReconciler) GetReconcileOrders

func (r *PytorchJobReconciler) GetReconcileOrders() []v1.ReplicaType

func (*PytorchJobReconciler) GetServicesForJob

func (r *PytorchJobReconciler) GetServicesForJob(obj interface{}) ([]*corev1.Service, error)

GetServicesForJob returns the services managed by the job. This can be achieved by selecting services using label key "job-name" i.e. all services created by the job will come with label "job-name" = <this_job_name>

func (*PytorchJobReconciler) IsMasterRole

func (r *PytorchJobReconciler) IsMasterRole(replicas map[v1.ReplicaType]*v1.ReplicaSpec, rtype v1.ReplicaType, index int) bool

IsMasterRole returns if this replica type with index specified is a master role. MasterRole pod will have "job-role=master" set in its label

func (*PytorchJobReconciler) Reconcile

func (r *PytorchJobReconciler) Reconcile(req ctrl.Request) (ctrl.Result, error)

func (*PytorchJobReconciler) SetClusterSpec

func (r *PytorchJobReconciler) SetClusterSpec(job interface{}, podTemplate *corev1.PodTemplateSpec, rtype, index string) error

SetClusterSpec sets the cluster spec for the pod

func (*PytorchJobReconciler) SetupWithManager

func (r *PytorchJobReconciler) SetupWithManager(mgr ctrl.Manager) error

func (*PytorchJobReconciler) UpdateJobStatus

func (r *PytorchJobReconciler) UpdateJobStatus(job interface{}, replicas map[v1.ReplicaType]*v1.ReplicaSpec, jobStatus *v1.JobStatus, restart bool) error

UpdateJobStatus updates the job status and job conditions

func (*PytorchJobReconciler) UpdateJobStatusInApiServer

func (r *PytorchJobReconciler) UpdateJobStatusInApiServer(job interface{}, jobStatus *v1.JobStatus) error

UpdateJobStatusInApiServer updates the job status in API server