Provide a metric for cohort resource reservations#9833
Open
mszadkow wants to merge 2 commits intokubernetes-sigs:mainfrom
Open
Provide a metric for cohort resource reservations#9833mszadkow wants to merge 2 commits intokubernetes-sigs:mainfrom
mszadkow wants to merge 2 commits intokubernetes-sigs:mainfrom
Conversation
k8s-ci-robot
added
release-note
Contributor
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: mszadkow The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
k8s-ci-robot
added
size/L
✅ Deploy Preview for kubernetes-sigs-kueue canceled.
|
k8s-ci-robot
added
the
needs-rebase
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
from
1cc9dc8 to
081ab40
Compare
k8s-ci-robot
added
needs-rebase
pajakd
reviewed
Mar 16, 2026
pajakd
reviewed
Mar 16, 2026
test/integration/singlecluster/scheduler/fairsharing/metrics_test.go
Outdated
Show resolved
Hide resolved
pajakd
reviewed
Mar 16, 2026
test/integration/singlecluster/scheduler/fairsharing/metrics_test.go
Outdated
Show resolved
Hide resolved
pajakd
reviewed
Mar 16, 2026
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
2 times, most recently
from
f87ae90 to
a26f57a
Compare
k8s-ci-robot
removed
the
needs-rebase
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
from
a26f57a to
c94a40d
Compare
vladikkuzn
reviewed
Mar 17, 2026
vladikkuzn
reviewed
Mar 17, 2026
Contributor
|
/lgtm |
k8s-ci-robot
added
the
lgtm
Contributor
|
LGTM label has been added. DetailsGit tree hash: 63f91d381ca482809071ea08ba34d1800d818d03 |
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
from
c94a40d to
35c178e
Compare
k8s-ci-robot
removed
the
lgtm
Contributor
|
New changes are detected. LGTM label has been removed. |
mbobrovskyi
reviewed
Mar 19, 2026
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
from
11cf83f to
8761860
Compare
Contributor
Author
|
@mbobrovskyi please check again |
mbobrovskyi
reviewed
Mar 20, 2026
mbobrovskyi
reviewed
Mar 20, 2026
mbobrovskyi
reviewed
Mar 20, 2026
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
from
8761860 to
b85a8be
Compare
mimowo
reviewed
Mar 20, 2026
mimowo
reviewed
Mar 20, 2026
mimowo
reviewed
Mar 20, 2026
mimowo
reviewed
Mar 20, 2026
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
from
b85a8be to
58b1de4
Compare
mszadkow
changed the title
mimowo
reviewed
Mar 20, 2026
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
from
58b1de4 to
06f08e6
Compare
mimowo
reviewed
Mar 20, 2026
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
2 times, most recently
from
35dee7e to
ff6a34b
Compare
mimowo
reviewed
Mar 20, 2026
Comment on lines
+68
to
+70
| // collectCohortMetricPoints prepares metric values for the target cohort and each ancestor. | ||
| // When simulateRemoval=true, it computes the remaining ancestor subtree values after | ||
| // subtracting the target cohort subtree quotas and reservations. |
Contributor
There was a problem hiding this comment.
Is it the intention?
// collectCohortMetricPoints prepares subtree metric points for the target cohort
// and all cohorts on the path from the target to the root.
// When simulateRemoval=true, it computes post-removal subtree values by subtracting
// the target cohort's subtree contribution from each cohort's current subtree totals.
// This is used when clearing metrics so ancestor subtree gauges are updated too,
// rather than left with stale values after the target cohort is removed.
Contributor
Author
There was a problem hiding this comment.
Yes, that's the point of this function
mimowo
reviewed
Mar 20, 2026
Comment on lines
83
to
120
| removedSubtreeQuota := ch.resourceNode.SubtreeQuota | ||
|
|
||
| // Cache subtree reservation aggregations once per cohort during this run to avoid | ||
| // repeated recursion when walking ancestor paths. | ||
| reservationsCache := make(map[*cohort]resources.FlavorResourceQuantities) | ||
| removedSubtreeReservations := totalSubtreeReservationsWithCache(ch, reservationsCache) | ||
|
|
||
| var points []cohortMetricPoint | ||
| for ancestor := range ch.PathSelfToRoot() { | ||
| quotas := ancestor.resourceNode.SubtreeQuota | ||
| cohortSubtreeQuota := ancestor.resourceNode.SubtreeQuota | ||
| cohortSubtreeReservations := totalSubtreeReservationsWithCache(ancestor, reservationsCache) | ||
|
|
||
| var ancestorCurrentSubtreeReservations resources.FlavorResourceQuantities | ||
| if simulateRemoval { | ||
| quotas = removedQuota | ||
| // In removal mode, subtract target subtree values from the ancestor current values | ||
| // to obtain the post-removal snapshot used for clearing metrics. | ||
| cohortSubtreeQuota = removedSubtreeQuota | ||
| cohortSubtreeReservations = removedSubtreeReservations | ||
| ancestorCurrentSubtreeReservations = totalSubtreeReservationsWithCache(ancestor, reservationsCache) | ||
| } | ||
|
|
||
| for flr, qty := range quotas { | ||
| flavorResourceKeys := sets.New[resources.FlavorResource]() | ||
| flavorResourceKeys.Insert(slices.Collect(maps.Keys(cohortSubtreeQuota))...) | ||
| flavorResourceKeys.Insert(slices.Collect(maps.Keys(cohortSubtreeReservations))...) | ||
|
|
||
| for fr := range flavorResourceKeys { | ||
| quotaQty := cohortSubtreeQuota[fr] | ||
| reservationsQty := cohortSubtreeReservations[fr] | ||
| if simulateRemoval { | ||
| qty = max(ancestor.resourceNode.SubtreeQuota[flr]-qty, 0) | ||
| quotaQty = max(ancestor.resourceNode.SubtreeQuota[fr]-quotaQty, 0) | ||
| reservationsQty = max(ancestorCurrentSubtreeReservations[fr]-reservationsQty, 0) | ||
| } | ||
| points = append(points, cohortMetricPoint{ | ||
| cohortName: ancestor.Name, | ||
| flavorResource: flr, | ||
| qty: qty, | ||
| cohortName: ancestor.Name, | ||
| flavorResource: fr, | ||
| quotaQty: quotaQty, | ||
| reservationsQty: reservationsQty, | ||
| }) |
Contributor
There was a problem hiding this comment.
The code is quite complex I would say. Maybe worth considering a helper to Subtract the usage maps represtented as ResourceFlavorQuantity maps to hide technicalities behind an abstraction and avoid some weird tricks with pinning / unpining the cohort variables.
I imagine the flow could be overall somthing like :
currentCohortQuota := ch.resourceNode.SubtreeQuota
for ancestor := range ch.PathSelfToRoot() {
ancestorCohortQuota := ancestor.resourceNode.SubtreeQuota
if simulateRemoval {
currentCohortQuota.Subtract(ancestorCohortQuota) // or currentCohortQuotaWithRemove = currentCohortQuota.Sub(ancestorCohortQuota) if we need to also keep previous value
}This would iiuc allow us later to skip the simulateRemoval handling in the flow.
Let me know if I missing something or your encounter some obstacles.
mimowo
reviewed
Mar 20, 2026
| } | ||
|
|
||
| removedQuota := ch.resourceNode.SubtreeQuota | ||
| removedSubtreeQuota := ch.resourceNode.SubtreeQuota |
Contributor
There was a problem hiding this comment.
Let's also add some unit tests for the function which would allow quick verification of correctness with debugger without running the heavy machinery of integration tests.
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
from
ff6a34b to
60b5ea1
Compare
k8s-ci-robot
added
the
needs-rebase
Contributor
|
PR needs rebase. DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
mszadkow
force-pushed
the
feat/7539-cohort-metrics-usage
branch
from
60b5ea1 to
fa2f796
Compare
k8s-ci-robot
added
size/XL
Contributor
|
@mszadkow: The following tests failed, say
Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR. DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What type of PR is this?
/kind feature
What this PR does / why we need it:
Add metrics for cohort's resource reservations.
Which issue(s) this PR fixes:
Relates to #7539
Special notes for your reviewer:
Does this PR introduce a user-facing change?