Skip to content

add docs accounting and resolve merge issue #554

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 1 commit into from
Aug 7, 2023
Merged

add docs accounting and resolve merge issue #554

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
11 changes: 8 additions & 3 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 @@ -718,6 +718,7 @@ func (qjm *XController) addTotalSnapshotResourcesConsumedByAw(totalgpu int32, to

func (qjm *XController) getAggregatedAvailableResourcesPriority(unallocatedClusterResources *clusterstateapi.
Resource, targetpr float64, requestingJob *arbv1.AppWrapper, agentId string) (*clusterstateapi.Resource, []*arbv1.AppWrapper) {
//get available free resources in the cluster.
r := unallocatedClusterResources.Clone()
// Track preemption resources
preemptable := clusterstateapi.EmptyResource()
Expand All @@ -732,7 +733,10 @@ func (qjm *XController) getAggregatedAvailableResourcesPriority(unallocatedClust
klog.Errorf("[getAggAvaiResPri] Unable to obtain the list of queueJobs %+v", err)
return r, nil
}

//for all AWs that have canRun status are true
//in non-preemption mode, we reserve resources for AWs
//reserving is done by subtracting total AW resources from pods owned by AW that are running or completed.
// AW can be running but items owned by it can be completed or there might be new set of pods yet to be spawned
for _, value := range queueJobs {
klog.V(10).Infof("[getAggAvaiResPri] %s: Evaluating job: %s to calculate aggregated resources.", time.Now().String(), value.Name)
if value.Name == requestingJob.Name {
Expand Down Expand Up @@ -797,10 +801,11 @@ func (qjm *XController) getAggregatedAvailableResourcesPriority(unallocatedClust

totalResource := qjm.addTotalSnapshotResourcesConsumedByAw(value.Status.TotalGPU, value.Status.TotalCPU, value.Status.TotalMemory)
klog.V(6).Infof("[getAggAvaiResPri] total resources consumed by Appwrapper %v when CanRun are %v", value.Name, totalResource)
pending, err = qjv.NonNegSub(totalResource)
delta, err := qjv.NonNegSub(totalResource)
pending = pending.Add(delta)
if err != nil {
klog.Warningf("[getAggAvaiResPri] Subtraction of resources failed, adding entire appwrapper resoources %v, %v", qjv, err)
pending = qjv
pending = pending.Add(qjv)
}
klog.V(6).Infof("[getAggAvaiResPri] The value of pending is %v", pending)
continue
Expand Down