feat(autoscaling): add burstable mode for workload autoscaler

[clamoriniere]

Summary

Introduces a burstable mode for workload autoscaling. When enabled, the controller applies CPU request recommendations but removes CPU limits from containers, allowing workloads to burst beyond their requested CPU when idle node capacity is available.

• Add alpha.autoscaling.datadoghq.com/burstable: "true" annotation support on DatadogPodAutoscaler: CPU limits are removed while CPU request recommendations are still applied
• Propagate the burstable annotation from a DatadogPodAutoscalerClusterProfile to all DPAs it generates; removing the annotation from the profile restores CPU limits on the next reconcile
• Rename annotation prefix from beta. to alpha. to better reflect the experimental nature of the feature

Test plan

• Unit tests added for controller_vertical_helpers (CPU limit removal logic)
• Unit tests added for pod_patcher (burstable recommendation ID suffix)
• Unit tests added for autoscaler_syncer (annotation propagation from DPACP to DPA)
• Unit tests added for controller (set/remove burstable annotation on K8s DPA object)
• Manual validation on kind cluster with DatadogPodAutoscalerClusterProfile carrying the annotation

🤖 Assisted-by: Claude Code Sonnet 4.6

[agent-platform-auto-pr]

Files inventory check summary

File checks results against ancestor 0f4a90a2:

Results for datadog-agent_7.79.0~devel.git.486.5b9d509.pipeline.107539161-1_amd64.deb:

No change detected

[agent-platform-auto-pr]

Static quality checks

✅ Please find below the results from static quality gates
Comparison made with ancestor 0f4a90a
📊 Static Quality Gates Dashboard
🔗 SQG Job

Successful checks

Info

	Quality gate	Change	Size (prev → curr → max)
✅	docker_agent_amd64	-10.27 KiB (0.00% reduction)	813.599 → 813.589 → 815.700
✅	docker_agent_arm64	-11.15 KiB (0.00% reduction)	816.772 → 816.761 → 821.970
✅	docker_agent_jmx_amd64	-10.26 KiB (0.00% reduction)	1004.514 → 1004.504 → 1006.580
✅	docker_agent_jmx_arm64	-11.16 KiB (0.00% reduction)	996.466 → 996.455 → 1001.570
✅	docker_cluster_agent_amd64	+8.52 KiB (0.00% increase)	204.111 → 204.119 → 206.270

25 successful checks with minimal change (< 2 KiB)

	Quality gate	Current Size
✅	agent_deb_amd64	753.327 MiB
✅	agent_deb_amd64_fips	710.235 MiB
✅	agent_heroku_amd64	313.466 MiB
✅	agent_rpm_amd64	753.310 MiB
✅	agent_rpm_amd64_fips	710.218 MiB
✅	agent_rpm_arm64	731.681 MiB
✅	agent_rpm_arm64_fips	691.649 MiB
✅	agent_suse_amd64	753.310 MiB
✅	agent_suse_amd64_fips	710.218 MiB
✅	agent_suse_arm64	731.681 MiB
✅	agent_suse_arm64_fips	691.649 MiB
✅	docker_cluster_agent_arm64	218.547 MiB
✅	docker_cws_instrumentation_amd64	7.142 MiB
✅	docker_cws_instrumentation_arm64	6.689 MiB
✅	docker_dogstatsd_amd64	39.270 MiB
✅	docker_dogstatsd_arm64	37.507 MiB
✅	dogstatsd_deb_amd64	29.917 MiB
✅	dogstatsd_deb_arm64	28.066 MiB
✅	dogstatsd_rpm_amd64	29.917 MiB
✅	dogstatsd_suse_amd64	29.917 MiB
✅	iot_agent_deb_amd64	43.305 MiB
✅	iot_agent_deb_arm64	40.352 MiB
✅	iot_agent_deb_armhf	41.100 MiB
✅	iot_agent_rpm_amd64	43.306 MiB
✅	iot_agent_suse_amd64	43.306 MiB

On-wire sizes (compressed)

	Quality gate	Change	Size (prev → curr → max)
✅	agent_deb_amd64	-35.01 KiB (0.02% reduction)	174.880 → 174.845 → 178.360
✅	agent_deb_amd64_fips	+25.04 KiB (0.01% increase)	165.396 → 165.420 → 172.790
✅	agent_heroku_amd64	neutral	75.042 MiB → 79.970
✅	agent_rpm_amd64	-36.7 KiB (0.02% reduction)	177.663 → 177.627 → 181.830
✅	agent_rpm_amd64_fips	-20.18 KiB (0.01% reduction)	167.765 → 167.746 → 173.370
✅	agent_rpm_arm64	-2.33 KiB (0.00% reduction)	159.612 → 159.610 → 163.060
✅	agent_rpm_arm64_fips	-14.84 KiB (0.01% reduction)	151.483 → 151.469 → 156.170
✅	agent_suse_amd64	-36.7 KiB (0.02% reduction)	177.663 → 177.627 → 181.830
✅	agent_suse_amd64_fips	-20.18 KiB (0.01% reduction)	167.765 → 167.746 → 173.370
✅	agent_suse_arm64	-2.33 KiB (0.00% reduction)	159.612 → 159.610 → 163.060
✅	agent_suse_arm64_fips	-14.84 KiB (0.01% reduction)	151.483 → 151.469 → 156.170
✅	docker_agent_amd64	+5.37 KiB (0.00% increase)	268.272 → 268.277 → 272.480
✅	docker_agent_arm64	+7.52 KiB (0.00% increase)	255.458 → 255.465 → 261.060
✅	docker_agent_jmx_amd64	+2.75 KiB (0.00% increase)	336.922 → 336.925 → 341.100
✅	docker_agent_jmx_arm64	+4.62 KiB (0.00% increase)	320.095 → 320.099 → 325.620
✅	docker_cluster_agent_amd64	+8.61 KiB (0.01% increase)	71.421 → 71.429 → 72.920
✅	docker_cluster_agent_arm64	+4.84 KiB (0.01% increase)	67.046 → 67.051 → 68.220
✅	docker_cws_instrumentation_amd64	neutral	2.999 MiB → 3.330
✅	docker_cws_instrumentation_arm64	neutral	2.729 MiB → 3.090
✅	docker_dogstatsd_amd64	neutral	15.180 MiB → 15.820
✅	docker_dogstatsd_arm64	neutral	14.506 MiB → 14.830
✅	dogstatsd_deb_amd64	neutral	7.899 MiB → 8.790
✅	dogstatsd_deb_arm64	neutral	6.784 MiB → 7.710
✅	dogstatsd_rpm_amd64	neutral	7.911 MiB → 8.800
✅	dogstatsd_suse_amd64	neutral	7.911 MiB → 8.800
✅	iot_agent_deb_amd64	-4.1 KiB (0.04% reduction)	11.411 → 11.407 → 13.040
✅	iot_agent_deb_arm64	neutral	9.726 MiB → 11.450
✅	iot_agent_deb_armhf	neutral	9.946 MiB → 11.620
✅	iot_agent_rpm_amd64	-4.79 KiB (0.04% reduction)	11.428 → 11.423 → 13.060
✅	iot_agent_suse_amd64	-4.79 KiB (0.04% reduction)	11.428 → 11.423 → 13.060

[cit-pr-commenter-54b7da]

Regression Detector

Regression Detector Results

Metrics dashboard
Target profiles
Run ID: 6fa2e19e-e230-4482-879d-29c398007ce7

Baseline: 0f4a90a
Comparison: 141d8c4
Diff

Optimization Goals: ✅ No significant changes detected

Experiments ignored for regressions

Regressions in experiments with settings containing erratic: true are ignored.

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	docker_containers_cpu	% cpu utilization	-0.69	[-3.73, +2.35]	1	Logs

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	tcp_syslog_to_blackhole	ingress throughput	+2.59	[+2.41, +2.77]	1	Logs
➖	ddot_metrics_sum_cumulativetodelta_exporter	memory utilization	+0.39	[+0.16, +0.61]	1	Logs
➖	otlp_ingest_logs	memory utilization	+0.13	[+0.02, +0.25]	1	Logs
➖	ddot_metrics_sum_delta	memory utilization	+0.07	[-0.10, +0.24]	1	Logs
➖	file_to_blackhole_500ms_latency	egress throughput	+0.06	[-0.34, +0.46]	1	Logs
➖	file_to_blackhole_1000ms_latency	egress throughput	+0.04	[-0.39, +0.47]	1	Logs
➖	quality_gate_idle_all_features	memory utilization	+0.02	[-0.01, +0.06]	1	Logs bounds checks dashboard
➖	tcp_dd_logs_filter_exclude	ingress throughput	+0.00	[-0.11, +0.11]	1	Logs
➖	uds_dogstatsd_to_api_v3	ingress throughput	-0.01	[-0.21, +0.20]	1	Logs
➖	uds_dogstatsd_20mb_12k_contexts_20_senders	memory utilization	-0.02	[-0.08, +0.04]	1	Logs
➖	uds_dogstatsd_to_api	ingress throughput	-0.04	[-0.25, +0.17]	1	Logs
➖	file_to_blackhole_0ms_latency	egress throughput	-0.05	[-0.56, +0.47]	1	Logs
➖	file_to_blackhole_100ms_latency	egress throughput	-0.06	[-0.19, +0.07]	1	Logs
➖	file_tree	memory utilization	-0.10	[-0.16, -0.05]	1	Logs
➖	quality_gate_idle	memory utilization	-0.17	[-0.22, -0.12]	1	Logs bounds checks dashboard
➖	docker_containers_memory	memory utilization	-0.30	[-0.39, -0.22]	1	Logs
➖	otlp_ingest_metrics	memory utilization	-0.31	[-0.46, -0.15]	1	Logs
➖	ddot_metrics	memory utilization	-0.39	[-0.57, -0.21]	1	Logs
➖	ddot_logs	memory utilization	-0.51	[-0.57, -0.45]	1	Logs
➖	ddot_metrics_sum_cumulative	memory utilization	-0.67	[-0.81, -0.52]	1	Logs
➖	docker_containers_cpu	% cpu utilization	-0.69	[-3.73, +2.35]	1	Logs
➖	quality_gate_metrics_logs	memory utilization	-1.54	[-1.77, -1.32]	1	Logs bounds checks dashboard
➖	quality_gate_logs	% cpu utilization	-1.78	[-3.40, -0.15]	1	Logs bounds checks dashboard

Bounds Checks: ✅ Passed

perf	experiment	bounds_check_name	replicates_passed	observed_value	links
✅	docker_containers_cpu	simple_check_run	10/10	707 ≥ 26
✅	docker_containers_memory	memory_usage	10/10	271.60MiB ≤ 370MiB
✅	docker_containers_memory	simple_check_run	10/10	707 ≥ 26
✅	file_to_blackhole_0ms_latency	memory_usage	10/10	0.19GiB ≤ 1.20GiB
✅	file_to_blackhole_0ms_latency	missed_bytes	10/10	0B = 0B
✅	file_to_blackhole_1000ms_latency	memory_usage	10/10	0.23GiB ≤ 1.20GiB
✅	file_to_blackhole_1000ms_latency	missed_bytes	10/10	0B = 0B
✅	file_to_blackhole_100ms_latency	memory_usage	10/10	0.19GiB ≤ 1.20GiB
✅	file_to_blackhole_100ms_latency	missed_bytes	10/10	0B = 0B
✅	file_to_blackhole_500ms_latency	memory_usage	10/10	0.21GiB ≤ 1.20GiB
✅	file_to_blackhole_500ms_latency	missed_bytes	10/10	0B = 0B
✅	quality_gate_idle	intake_connections	10/10	3 = 3	bounds checks dashboard
✅	quality_gate_idle	memory_usage	10/10	171.85MiB ≤ 181MiB	bounds checks dashboard
✅	quality_gate_idle_all_features	intake_connections	10/10	3 = 3	bounds checks dashboard
✅	quality_gate_idle_all_features	memory_usage	10/10	488.70MiB ≤ 550MiB	bounds checks dashboard
✅	quality_gate_logs	intake_connections	10/10	3 ≤ 6	bounds checks dashboard
✅	quality_gate_logs	memory_usage	10/10	206.46MiB ≤ 220MiB	bounds checks dashboard
✅	quality_gate_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_metrics_logs	cpu_usage	10/10	359.53 ≤ 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	408.84MiB ≤ 475MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	missed_bytes	10/10	0B = 0B	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".

CI Pass/Fail Decision

✅ Passed. All Quality Gates passed.

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