Versions in this module Expand all Collapse all v0 v0.2.0 Apr 8, 2020 Changes in this version + type EtcdClient struct + func NewEtcdClient(endpoints string, numPservers int, dialtimeout time.Duration, ttlSec int) *EtcdClient + func (e *EtcdClient) GetKey(key string, timeout time.Duration) ([]byte, error) + func (e *EtcdClient) PutKey(key string, value []byte, timeout time.Duration, withLease bool) error + func (e *EtcdClient) Register(port int) error + func (e *EtcdClient) Shutdown() error v0.1.0 Jan 12, 2020 Changes in this version + func AddResourceList(a v1.ResourceList, b v1.ResourceList) + type Autoscaler struct + func (a *Autoscaler) OnAdd(trainingjob *edlresource.TrainingJob) + func (a *Autoscaler) OnDel(trainingjob *edlresource.TrainingJob) + func (a *Autoscaler) OnUpdate(trainingjob *edlresource.TrainingJob) + func (a *Autoscaler) Run() + type Cluster struct + func (c *Cluster) CreateJob(j *batchv1.Job) (*batchv1.Job, error) + func (c *Cluster) CreateReplicaSet(r *v1beta1.ReplicaSet) (*v1beta1.ReplicaSet, error) + func (c *Cluster) DeleteReplicaSet(namespace, name string) error + func (c *Cluster) DeleteTrainerJob(namespace, name string) error + func (c *Cluster) GetReplicaSet(namespace, name string) (*v1beta1.ReplicaSet, error) + func (c *Cluster) InquiryResource() (res ClusterResource, err error) + func (c Cluster) GetTrainerJob(job *edlresource.TrainingJob) (*batchv1.Job, error) + func (c Cluster) GetTrainerJobByName(namespace, name string) (*batchv1.Job, error) + func (c Cluster) JobPods(job *edlresource.TrainingJob) (total, running, pending int, err error) + func (c Cluster) UpdateTrainerJob(job *batchv1.Job) error + type ClusterResource struct + CPULimitMilli int64 + CPURequestMilli int64 + CPUTotalMilli int64 + GPULimit int + GPURequest int + GPUTotal int + MemoryLimitMega int64 + MemoryRequestMega int64 + MemoryTotalMega int64 + NodeCount int + Nodes Nodes + type Controller struct + func New(c *rest.RESTClient, cs *kubernetes.Clientset, maxLoadDesired float64) (*Controller, error) + func (c *Controller) Run() + func (c *Controller) WatchTrainingJobs() + type DefaultJobParser int + func (p *DefaultJobParser) ParseToMaster(job *edlresource.TrainingJob) *v1beta1.ReplicaSet + func (p *DefaultJobParser) ParseToPserver(job *edlresource.TrainingJob) *v1beta1.ReplicaSet + func (p *DefaultJobParser) ParseToTrainer(job *edlresource.TrainingJob) *batchv1.Job + func (p *DefaultJobParser) Validate(job *edlresource.TrainingJob) error + type JobParser interface + ParseToMaster func(job *edlresource.TrainingJob) *v1beta1.ReplicaSet + ParseToPserver func(job *edlresource.TrainingJob) *v1beta1.ReplicaSet + ParseToTrainer func(job *edlresource.TrainingJob) *batchv1.Job + Validate func(job *edlresource.TrainingJob) error + type Nodes struct + NodesCPUIdleMilli map[string]int64 + NodesMemoryFreeMega map[string]int64 + func (ns *Nodes) String() string