Skip to content
This repository was archived by the owner on Sep 19, 2022. It is now read-only.

use the priority of kube-batch #209

Open
wants to merge 13 commits into
base: master
Choose a base branch
from
+51 −5
Open

use the priority of kube-batch #209

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
9b5cdab
add priority
YesterdayxD Aug 19, 2019
e3b3ea0
add priority on kube-batch PriorityClassName
YesterdayxD Aug 19, 2019
aeb093e
add English comment and add priority in SyncPodGroup function
YesterdayxD Aug 20, 2019
09d6c1b
merge develop
YesterdayxD Aug 20, 2019
ba553a4
add priority in v1beta2
YesterdayxD Aug 20, 2019
a50a725
Merge branch 'develop'
YesterdayxD Aug 20, 2019
6e17f2c
debug undefined: priorityClassName in v1beta2
YesterdayxD Aug 20, 2019
3683cfa
Merge branch 'develop'
YesterdayxD Aug 20, 2019
47fed48
add priority in zz_generated.deepcopy.go
YesterdayxD Aug 20, 2019
a49a402
Merge branch 'develop'
YesterdayxD Aug 20, 2019
aad3b4c
recover origin code
YesterdayxD Aug 20, 2019
2442fdc
fix goimports
YesterdayxD Aug 22, 2019
6c24d6e
fix gometalinter about merge variable declaration with assignment on …
YesterdayxD Aug 22, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions manifests/crd.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,8 @@ spec:
properties:
spec:
properties:
priorityClassName:
type: string
pytorchReplicaSpecs:
properties:
Master:
Expand Down
2 changes: 2 additions & 0 deletions manifests/podgroup.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,8 @@ spec:
minMember:
format: int32
type: integer
priorityClassName:
type: string
type: object
status:
properties:
Expand Down
4 changes: 4 additions & 0 deletions pkg/apis/pytorch/v1/types.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,6 +69,10 @@ type PyTorchJobSpec struct {
// "Worker": PyTorchReplicaSpec,
// }
PyTorchReplicaSpecs map[PyTorchReplicaType]*common.ReplicaSpec `json:"pytorchReplicaSpecs"`

//add PriorityClassName
//PriorityClassName is a type of k8s resource.(kubectl get priorityclass)
PriorityClassName *string `json:"priorityClassName,omitempty"`
}

// PyTorchReplicaType is the type for PyTorchReplica. Can be one of "Master" or "Worker".
Expand Down
5 changes: 5 additions & 0 deletions pkg/apis/pytorch/v1/zz_generated.deepcopy.go
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

4 changes: 4 additions & 0 deletions pkg/apis/pytorch/v1beta2/types.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,6 +72,10 @@ type PyTorchJobSpec struct {
// "Worker": PyTorchReplicaSpec,
// }
PyTorchReplicaSpecs map[PyTorchReplicaType]*common.ReplicaSpec `json:"pytorchReplicaSpecs"`

//add PriorityClassName
//PriorityClassName is a type of k8s resource.(kubectl get priorityclass)
PriorityClassName *string `json:"priorityClassName,omitempty"`
}

// PyTorchReplicaType is the type for PyTorchReplica.
Expand Down
5 changes: 5 additions & 0 deletions pkg/apis/pytorch/v1beta2/zz_generated.deepcopy.go
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

6 changes: 4 additions & 2 deletions pkg/controller.v1/pytorch/controller.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,7 @@ import (

kubebatchclient "github.com/kubernetes-sigs/kube-batch/pkg/client/clientset/versioned"
log "github.com/sirupsen/logrus"
"k8s.io/api/core/v1"
v1 "k8s.io/api/core/v1"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
"k8s.io/apimachinery/pkg/runtime/schema"
utilruntime "k8s.io/apimachinery/pkg/util/runtime"
Expand Down Expand Up @@ -437,7 +437,9 @@ func (pc *PyTorchController) reconcilePyTorchJobs(job *pyv1.PyTorchJob) error {

if pc.Config.EnableGangScheduling {
minAvailableReplicas := getTotalReplicas(job)
_, err := pc.SyncPodGroup(job, minAvailableReplicas)
priorityClassName := getPriorityClassName(job)
//_, err := pc.SyncPodGroup(job, minAvailableReplicas)
_, err := pc.SyncPodGroup(job, minAvailableReplicas, priorityClassName)
if err != nil {
logger.Warnf("Sync PodGroup %v: %v", job.Name, err)
}
Expand Down
5 changes: 5 additions & 0 deletions pkg/controller.v1/pytorch/job.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -216,3 +216,8 @@ func getTotalFailedReplicas(job *pyv1.PyTorchJob) int32 {
}
return totalFailedReplicas
}

func getPriorityClassName(job *pyv1.PyTorchJob) string {
priorityClassName := *(job.Spec.PriorityClassName)
Copy link
Member

@gaocegege gaocegege Aug 23, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There may be an runtime error: invalid memory address or nil pointer dereference

Copy link
Author

@YesterdayxD YesterdayxD Aug 23, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I need to remove the bracket

Copy link
Author

@YesterdayxD YesterdayxD Aug 23, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

right？ @gaocegege

Copy link
Member

@gaocegege gaocegege Aug 23, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No I think you need to check if it is nil.

Copy link
Author

@YesterdayxD YesterdayxD Aug 23, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I got it.

return priorityClassName
}
1 change: 1 addition & 0 deletions pkg/controller.v1/pytorch/status.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,6 +69,7 @@ func (pc *PyTorchController) updateStatusSingle(job *pyv1.PyTorchJob, rtype pyv1

// Expect to have `replicas - succeeded` pods alive.
commonType := common.ReplicaType(rtype)
//expected is a flag of success.if expected==0,PyTorchJob is successfully completed.
expected := replicas - int(job.Status.ReplicaStatuses[commonType].Succeeded)
running := int(job.Status.ReplicaStatuses[commonType].Active)
failed := int(job.Status.ReplicaStatuses[commonType].Failed)
Expand Down
5 changes: 3 additions & 2 deletions pkg/controller.v1beta2/pytorch/controller.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,7 @@ import (

kubebatchclient "github.com/kubernetes-sigs/kube-batch/pkg/client/clientset/versioned"
log "github.com/sirupsen/logrus"
"k8s.io/api/core/v1"
v1 "k8s.io/api/core/v1"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
"k8s.io/apimachinery/pkg/runtime/schema"
utilruntime "k8s.io/apimachinery/pkg/util/runtime"
Expand Down Expand Up @@ -428,7 +428,8 @@ func (pc *PyTorchController) reconcilePyTorchJobs(job *v1beta2.PyTorchJob) error

if pc.Config.EnableGangScheduling {
minAvailableReplicas := getTotalReplicas(job)
_, err := pc.SyncPodGroup(job, minAvailableReplicas)
priorityClassName := getPriorityClassName(job)
_, err := pc.SyncPodGroup(job, minAvailableReplicas, priorityClassName)
if err != nil {
logger.Warnf("Sync PodGroup %v: %v", job.Name, err)
}
Expand Down
5 changes: 5 additions & 0 deletions pkg/controller.v1beta2/pytorch/job.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -206,3 +206,8 @@ func getTotalFailedReplicas(job *v1beta2.PyTorchJob) int32 {
}
return totalFailedReplicas
}

func getPriorityClassName(job *v1beta2.PyTorchJob) string {
priorityClassName := *(job.Spec.PriorityClassName)
return priorityClassName
}
3 changes: 2 additions & 1 deletion vendor/github.com/kubeflow/tf-operator/pkg/common/jobcontroller/jobcontroller.go
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

9 changes: 9 additions & 0 deletions vendor/github.com/kubernetes-sigs/kube-batch/pkg/apis/scheduling/v1alpha1/types.go
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.