Skip to content

simplify acc logic #451

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Jun 30, 2023
Merged

simplify acc logic #451

Changes from 1 commit
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
25166ce
simply acc logic
asm582 Jun 30, 2023
86ae9e3
fix test
asm582 Jun 30, 2023
31277eb
change log level
asm582 Jun 30, 2023
874a076
address review
asm582 Jun 30, 2023
edd192b
resolve review-2
asm582 Jun 30, 2023
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
15 changes: 9 additions & 6 deletions pkg/controller/queuejob/queuejob_controller_ex.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -908,7 +908,7 @@ func (qjm *XController) getAggregatedAvailableResourcesPriority(unallocatedClust
totalResource := qjm.addTotalSnapshotResourcesConsumedByAw(value.Status.TotalGPU, value.Status.TotalCPU, value.Status.TotalMemory)
preemptable = preemptable.Add(totalResource)
continue
} else if value.Status.SystemPriority < targetpr || !value.Status.CanRun {
} else if value.Status.SystemPriority < targetpr {
// Dispatcher Mode: Ensure this job is part of the target cluster
if qjm.isDispatcher {
// Get the job key
Expand Down Expand Up @@ -940,23 +940,26 @@ func (qjm *XController) getAggregatedAvailableResourcesPriority(unallocatedClust
klog.V(10).Infof("[getAggAvaiResPri] %s: Skipping adjustments for %s since priority %f is >= %f of requesting job: %s.", time.Now().String(),
value.Name, value.Status.SystemPriority, targetpr, requestingJob.Name)
continue
} else if value.Status.CanRun && (targetpr < value.Status.SystemPriority) {
} else if value.Status.CanRun {
qjv := clusterstateapi.EmptyResource()
for _, resctrl := range qjm.qjobResControls {
qjv = resctrl.GetAggregatedResources(value)
res := resctrl.GetAggregatedResources(value)
qjv.Add(res)
klog.V(10).Infof("[getAggAvaiResPri] Subtract all resources %+v in resctrlType=%T for job %s which can-run is set to: %v but state is still pending.", qjv, resctrl, value.Name, value.Status.CanRun)
}
for _, genericItem := range value.Spec.AggrResources.GenericItems {
qjv, _ = genericresource.GetResources(&genericItem)
res, _ := genericresource.GetResources(&genericItem)
qjv.Add(res)
klog.V(10).Infof("[getAggAvaiResPri] Subtract all resources %+v in resctrlType=%T for job %s which can-run is set to: %v but state is still pending.", qjv, genericItem, value.Name, value.Status.CanRun)
}

totalResource := qjm.addTotalSnapshotResourcesConsumedByAw(value.Status.TotalGPU, value.Status.TotalCPU, value.Status.TotalMemory)
pending, err = qjv.NonNegSub(totalResource)
delta, err := qjv.NonNegSub(totalResource)
if err != nil {
klog.Errorf("[getAggAvaiResPri] Subtraction of resources failed, adding entire appwrapper resoources %v", qjv)
klog.Errorf("[getAggAvaiResPri] Subtraction of resources failed, adding entire appwrapper resoources %v, %v", qjv, err)
pending = qjv
}
pending = pending.Add(delta)
continue
} else {
//Do nothing
Expand Down