feat(autoscaling): CPURequestsRemoveLimitsMemoryRequestsAndLimits + cluster burstable default

[clamoriniere]

Summary

1. CPURequestsRemoveLimitsMemoryRequestsAndLimits controlled value

Implements the new CPURequestsRemoveLimitsMemoryRequestsAndLimits container controlledValues enum from the datadog-operator (c59fc90).

When set on a container constraint:

• CPU: only requests are controlled; any existing CPU limit is actively removed from the live pod, letting the container burst freely and avoid CPU throttling.
• Memory: both requests and limits are controlled (standard RequestsAndLimits semantics).

The removal flows via a sentinel quantity (-1) stored on ContainerResources.Limits[cpu] by applyVerticalConstraints and recognised by the pod patcher, which actively deletes any pre-existing CPU limit from the live pod (an absent entry would mean "leave untouched").

2. Recognise spec.options.burstable in IsBurstable()

IsBurstable() now resolves from spec.options.burstable first, then the existing preview annotation. There is no cluster-level default — burstable mode requires an explicit per-DPA opt-in via either signal.

3. Report effective burstable mode in status.options.burstable

BuildStatus populates status.options.burstable so consumers can observe the resolved value:

• When spec.options.burstable is set: reported as-is (true or false).
• When the preview annotation enables burstable: reported as true.
• Otherwise: omitted (no opt-in).

4. Fix stale-sentinel propagation when the backend stops emitting a vertical recommendation

SetActiveScalingValues previously self-assigned p.scalingValues.Vertical when the active vertical source was nil (because selectScalingValues(nil) returns p.scalingValues). This kept any burstable sentinel from the previous cycle alive and caused applyVerticalConstraints(burstable=false) to early-return, suppressing the rollout when burstable was toggled off. Now Vertical is explicitly reset to nil so the next sync sees "no recommendation" and clears live state.

5. Dependency bump

go.mod: datadog-operator/api → v0.0.0-20260515125012-8e158b708444.

Test plan

• TestApplyVerticalConstraints_CPURequestsRemoveLimits — CPU limit stripped from recommendation, memory limit preserved, sentinel inserted.
• TestPatchContainerResources_* / TestPatchPod_* — CPU limit actively removed from live pod, idempotent when already absent.
• IsBurstable priority matrix — spec > annotation > default false.
• BuildStatus reports options.burstable for spec (true/false) and for annotation-enabled, omits when neither is set.
• TestSetActiveScalingValues_NilSource_ClearsVertical — nil vertical source clears scalingValues.Vertical instead of self-assigning the stale sentinel.
• Full package suite passes: dda inv test --targets=./pkg/clusteragent/autoscaling/workload/...
• Lint clean: dda inv linter.go --targets ./pkg/clusteragent/autoscaling/workload/...
• Manual QA in kind cluster: DPA with preview annotation reports status.options.burstable: true; DPA without explicit opt-in reports no options (defaults to non-burstable).

🤖 PR description and code assisted by Claude:claude-opus-4-7

[dd-octo-sts]

Files inventory check summary

File checks results against ancestor efc625c8:

Results for datadog-agent_7.81.0~devel.git.587.bebf5b2.pipeline.117750425-1_amd64.deb:

No change detected

[cit-pr-commenter-54b7da]

Regression Detector

Regression Detector Results

Metrics dashboard
Target profiles
Run ID: 0b50317c-c1db-4bdf-b93e-65f955aedede

Baseline: efc625c
Comparison: bebf5b2
Diff

Optimization Goals: ✅ No significant changes detected

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	quality_gate_metrics_logs	memory utilization	+1.67	[+1.41, +1.93]	1	Logs bounds checks dashboard
➖	quality_gate_idle	memory utilization	+0.53	[+0.47, +0.59]	1	Logs bounds checks dashboard
➖	quality_gate_idle_all_features	memory utilization	-0.44	[-0.48, -0.40]	1	Logs bounds checks dashboard
➖	quality_gate_logs	% cpu utilization	-1.86	[-2.88, -0.83]	1	Logs bounds checks dashboard

Bounds Checks: ✅ Passed

perf	experiment	bounds_check_name	replicates_passed	observed_value	links
✅	quality_gate_idle	intake_connections	10/10	3 ≤ 4	bounds checks dashboard
✅	quality_gate_idle	memory_usage	10/10	146.51MiB ≤ 147MiB	bounds checks dashboard
✅	quality_gate_idle	total_bytes_received	10/10	729.93KiB ≤ 819.20KiB	bounds checks dashboard
✅	quality_gate_idle_all_features	intake_connections	10/10	3 ≤ 4	bounds checks dashboard
✅	quality_gate_idle_all_features	memory_usage	10/10	476.84MiB ≤ 495MiB	bounds checks dashboard
✅	quality_gate_idle_all_features	total_bytes_received	10/10	1.13MiB ≤ 1.25MiB	bounds checks dashboard
✅	quality_gate_logs	intake_connections	10/10	3 ≤ 6	bounds checks dashboard
✅	quality_gate_logs	memory_usage	10/10	178.74MiB ≤ 195MiB	bounds checks dashboard
✅	quality_gate_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_logs	total_bytes_received	10/10	264.02MiB ≤ 292MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	cpu_usage	10/10	342.14 ≤ 2000	bounds checks dashboard
✅	quality_gate_metrics_logs	intake_connections	10/10	4 ≤ 6	bounds checks dashboard
✅	quality_gate_metrics_logs	memory_usage	10/10	395.55MiB ≤ 430MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_metrics_logs	total_bytes_received	10/10	0.94GiB ≤ 1.04GiB	bounds checks dashboard

Explanation

Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%

Performance changes are noted in the perf column of each table:

• ✅ = significantly better comparison variant performance
• ❌ = significantly worse comparison variant performance
• ➖ = no significant change in performance

A regression test is an A/B test of target performance in a repeatable rig, where "performance" is measured as "comparison variant minus baseline variant" for an optimization goal (e.g., ingress throughput). Due to intrinsic variability in measuring that goal, we can only estimate its mean value for each experiment; we report uncertainty in that value as a 90.00% confidence interval denoted "Δ mean % CI".

For each experiment, we decide whether a change in performance is a "regression" -- a change worth investigating further -- if all of the following criteria are true:

1. Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.

2. Its 90.00% confidence interval "Δ mean % CI" does not contain zero, indicating that if our statistical model is accurate, there is at least a 90.00% chance there is a difference in performance between baseline and comparison variants.

3. Its configuration does not mark it "erratic".

Replicate Execution Details

We run multiple replicates for each experiment/variant. However, we allow replicates to be automatically retried if there are any failures, up to 8 times, at which point the replicate is marked dead and we are unable to run analysis for the entire experiment. We call each of these attempts at running replicates a replicate execution. This section lists all replicate executions that failed due to the target crashing or being oom killed.

Note: In the below tables we bucket failures by experiment, variant, and failure type. For each of these buckets we list out the replicate indexes that failed with an annotation signifying how many times said replicate failed with the given failure mode. In the below example the baseline variant of the experiment named experiment_with_failures had two replicates that failed by oom kills. Replicate 0, which failed 8 executions, and replicate 1 which failed 6 executions, all with the same failure mode.

Experiment	Variant	Replicates	Failure	Logs	Debug Dashboard
experiment_with_failures	baseline	0 (x8) 1 (x6)	Oom killed		Debug Dashboard

The debug dashboard links will take you to a debugging dashboard specifically designed to investigate replicate execution failures.

❌ Retried Profiling Replicate Execution Failures (ddprof)

Note: Profiling replicas may still be executing. See the debug dashboard for up to date status.

Experiment	Variant	Replicates	Failure	Debug Dashboard
quality_gate_idle	baseline	10	Oom killed	Debug Dashboard
quality_gate_idle_all_features	baseline	10	Oom killed	Debug Dashboard
quality_gate_idle_all_features	comparison	10	Oom killed	Debug Dashboard
quality_gate_logs	baseline	10	Oom killed	Debug Dashboard
quality_gate_logs	comparison	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	baseline	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	comparison	10	Oom killed	Debug Dashboard

CI Pass/Fail Decision

✅ Passed. All Quality Gates passed.

• quality_gate_idle, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check cpu_usage: 10/10 replicas passed. Gate passed.

[dd-octo-sts]

This pull request has been automatically marked as stale because it has not had activity in the past 15 days.

It will be closed in 30 days if no further activity occurs. If this pull request is still relevant, adding a comment or pushing new commits will keep it open. Also, you can always reopen the pull request if you missed the window.

Thank you for your contributions!

[datadog-official]

🎯 Code Coverage (details)
• Patch Coverage: 86.67%
• Overall Coverage: 50.58% (-0.00%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: bebf5b2 | Docs | Datadog PR Page | Give us feedback!}

[dd-octo-sts]

Static quality checks

✅ Please find below the results from static quality gates
Comparison made with ancestor efc625c
📊 Static Quality Gates Dashboard
🔗 SQG Job

Successful checks

Info

	Quality gate	Change	Size (prev → curr → max)
✅	docker_agent_amd64	+5.41 KiB (0.00% increase, -0.57% of buffer)	812.857 → 812.863 → 813.790
✅	docker_agent_jmx_amd64	+5.41 KiB (0.00% increase, -0.70% of buffer)	1003.798 → 1003.804 → 1004.550
✅	docker_cluster_agent_amd64	+4.29 KiB (0.00% increase, -0.58% of buffer)	209.741 → 209.746 → 210.470
✅	docker_host_profiler_amd64	+6.62 KiB (0.00% increase, -0.05% of buffer)	304.503 → 304.509 → 317.640

28 successful checks with minimal change (< 2 KiB)

	Quality gate	Current Size
✅	agent_deb_amd64	753.426 MiB
✅	agent_deb_amd64_fips	709.201 MiB
✅	agent_heroku_amd64	312.154 MiB
✅	agent_rpm_amd64	753.410 MiB
✅	agent_rpm_amd64_fips	709.185 MiB
✅	agent_rpm_arm64	728.713 MiB
✅	agent_rpm_arm64_fips	688.228 MiB
✅	agent_suse_amd64	753.410 MiB
✅	agent_suse_amd64_fips	709.185 MiB
✅	agent_suse_arm64	728.713 MiB
✅	agent_suse_arm64_fips	688.228 MiB
✅	docker_agent_arm64	813.187 MiB
✅	docker_agent_jmx_arm64	992.780 MiB
✅	docker_cluster_agent_arm64	222.847 MiB
✅	docker_cws_instrumentation_amd64	7.447 MiB
✅	docker_cws_instrumentation_arm64	6.877 MiB
✅	docker_dogstatsd_amd64	39.833 MiB
✅	docker_dogstatsd_arm64	37.883 MiB
✅	docker_host_profiler_arm64	315.633 MiB
✅	dogstatsd_deb_amd64	30.490 MiB
✅	dogstatsd_deb_arm64	28.487 MiB
✅	dogstatsd_rpm_amd64	30.490 MiB
✅	dogstatsd_suse_amd64	30.490 MiB
✅	iot_agent_deb_amd64	45.627 MiB
✅	iot_agent_deb_arm64	42.346 MiB
✅	iot_agent_deb_armhf	43.141 MiB
✅	iot_agent_rpm_amd64	45.628 MiB
✅	iot_agent_suse_amd64	45.627 MiB

[s-alad]

lgtm for agent config file

[dd-octo-sts]

This pull request has been automatically marked as stale because it has not had activity in the past 15 days.

It will be closed in 30 days if no further activity occurs. If this pull request is still relevant, adding a comment or pushing new commits will keep it open. Also, you can always reopen the pull request if you missed the window.

Thank you for your contributions!

[LaurieScs]

LGTM, even though I'm not an agent expert 😅

[clamoriniere]

Hi @vboulineau, thanks for the review.

Following your feedback I scoped the PR down significantly. Since we decided not to enable the burstable option by default at the cluster level anymore, the global config setting, the PodAutoscalerInternalBuilder, and the podsGuaranteedQOS field on PodAutoscalerInternal were all dead weight — they're now gone. IsBurstable() is back to a parameter-less method that resolves from spec.options.burstable then the preview annotation, with no per-DPA cluster-level state.

I also walked back the unnecessary cosmetic changes (function renames, blank lines, reworded doc comments) so the diff stays focused on the actual new functionality.

Ready for another round when you have time — hopefully closer to what you had in mind this time.