[CWS] add cleanup loop, removing persisted dumps that are no longer needed

[paulcacheux]

What does this PR do?

This PR adds a new cleanup goroutine, that will remove the unneeded activity dumps from the directory managed by the directory provider.

The basic rule used is that if there is a real profile (i.e. tag == "") for a given image name, a dump (tag != "") is not needed (and we already have the logic to not even load it).

The main goal is to drastically cut on the amount of calls to LoadProfile that are done, reducing clearly the amount of allocations (obviously since we load way less profiles).

Motivation

Additional Notes

Possible Drawbacks / Trade-offs

Describe how to test/QA your changes

[pr-commenter]

Regression Detector

Regression Detector Results

Metrics dashboard
Target profiles
Run ID: 1ef06f16-0e98-4260-8ab7-457ed134b8f3

Baseline: d0a1d25
Comparison: 9fff824
Diff

Optimization Goals: ✅ No significant changes detected

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	uds_dogstatsd_to_api_cpu	% cpu utilization	+1.19	[+0.46, +1.92]	1	Logs
➖	basic_py_check	% cpu utilization	+0.97	[-2.85, +4.79]	1	Logs
➖	otel_to_otel_logs	ingress throughput	+0.65	[-0.06, +1.35]	1	Logs
➖	quality_gate_idle	memory utilization	+0.21	[+0.16, +0.25]	1	Logs bounds checks dashboard
➖	file_to_blackhole_300ms_latency	egress throughput	+0.11	[-0.52, +0.75]	1	Logs
➖	file_to_blackhole_100ms_latency	egress throughput	+0.09	[-0.59, +0.76]	1	Logs
➖	tcp_dd_logs_filter_exclude	ingress throughput	+0.00	[-0.01, +0.01]	1	Logs
➖	uds_dogstatsd_to_api	ingress throughput	-0.02	[-0.12, +0.08]	1	Logs
➖	file_to_blackhole_500ms_latency	egress throughput	-0.05	[-0.82, +0.71]	1	Logs
➖	file_to_blackhole_0ms_latency	egress throughput	-0.08	[-0.94, +0.78]	1	Logs
➖	quality_gate_idle_all_features	memory utilization	-0.13	[-0.24, -0.02]	1	Logs bounds checks dashboard
➖	tcp_syslog_to_blackhole	ingress throughput	-0.23	[-0.30, -0.16]	1	Logs
➖	file_to_blackhole_1000ms_latency	egress throughput	-0.31	[-1.09, +0.47]	1	Logs
➖	file_to_blackhole_1000ms_latency_linear_load	egress throughput	-0.36	[-0.81, +0.10]	1	Logs
➖	file_tree	memory utilization	-0.43	[-0.57, -0.29]	1	Logs
➖	pycheck_lots_of_tags	% cpu utilization	-3.48	[-6.81, -0.15]	1	Logs

Bounds Checks: ❌ Failed

perf	experiment	bounds_check_name	replicates_passed	links
❌	file_to_blackhole_1000ms_latency	lost_bytes	0/10
✅	file_to_blackhole_0ms_latency	lost_bytes	10/10
✅	file_to_blackhole_0ms_latency	memory_usage	10/10
✅	file_to_blackhole_1000ms_latency	memory_usage	10/10
✅	file_to_blackhole_1000ms_latency_linear_load	memory_usage	10/10
✅	file_to_blackhole_100ms_latency	lost_bytes	10/10
✅	file_to_blackhole_100ms_latency	memory_usage	10/10
✅	file_to_blackhole_300ms_latency	lost_bytes	10/10
✅	file_to_blackhole_300ms_latency	memory_usage	10/10
✅	file_to_blackhole_500ms_latency	lost_bytes	10/10
✅	file_to_blackhole_500ms_latency	memory_usage	10/10
✅	quality_gate_idle	memory_usage	10/10	bounds checks dashboard
✅	quality_gate_idle_all_features	memory_usage	10/10	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_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.

[agent-platform-auto-pr]

[Fast Unit Tests Report]

On pipeline 49247697 (CI Visibility). The following jobs did not run any unit tests:

Jobs:

• tests_windows-x64

If you modified Go files and expected unit tests to run in these jobs, please double check the job logs. If you think tests should have been executed reach out to #agent-devx-help

[pr-commenter]

Test changes on VM

Use this command from test-infra-definitions to manually test this PR changes on a VM:

inv create-vm --pipeline-id=49247697 --os-family=ubuntu

Note: This applies to commit 9fff824

[spikat]

This feels like the cached profileMapping list is not very useful to me.. I wonder if we could modify the cached one to store "real" workload selector (instead of, somehow a "fake" profile one) and use it to clean up dumps more easily (instead of reconstruct the whole view by re-open every dumps/profiles every 5min). WDYT ?

[paulcacheux]

yes, good point. I updated the PR to cleanup a bit the profileMapping, and used it in the cleanup loop (instead of re-reading from disk)