Update the zstd for metrics default compression level to 1

[StephenWakely]

What does this PR do?

This updates the default compression level for compressing metric payloads with zstd to level 1.

Note, the default compression kind is still zlib, so this will only have an impact if people intentionally configure zstd.

Motivation

The default level of 5 provides a far superior compression ratio. It does this by allocating a chain table to track duplicate text. This does also require that significantly more memory is allocated. Whilst the memory is freed right after compression, we have noticed that with the glibc allocator this leads to significantly more memory fragmentation, especially at higher thread counts.

A longer term fix is to change the default allocator that the agent uses, however short term, dropping the default level to 1 provides a better compression ratio over zlib whilst not risking affecting RSS significantly.

Additional Notes

Possible Drawbacks / Trade-offs

Whilst still being better than zlib compression, this does reduce the compression ratio significantly. We've observed 40x on metric payloads (compression level 5 gives 60x).

Describe how to test/QA your changes

Configure the agent with

serializer_compressor_kind: zstd

Do not configure serializer_zstd_compressor_level.

[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=44059879 --os-family=ubuntu

Note: This applies to commit 4f3d78d

[StephenWakely]

/merge

[dd-devflow]

🚂 MergeQueue: waiting for PR to be ready

This merge request is not mergeable yet, because of pending checks/missing approvals. It will be added to the queue as soon as checks pass and/or get approvals.
Note: if you pushed new commits since the last approval, you may need additional approval.
You can remove it from the waiting list with /remove command.

Use /merge -c to cancel this operation!

[pr-commenter]

Regression Detector

Regression Detector Results

Run ID: 62026e0a-2621-475a-a34e-ce7c6ac82aa5 Metrics dashboard Target profiles

Baseline: 4a122d2
Comparison: 4f3d78d

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

No significant changes in experiment optimization goals

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

There were no significant changes in experiment optimization goals at this confidence level and effect size tolerance.

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	uds_dogstatsd_to_api_cpu	% cpu utilization	+0.64	[-0.12, +1.40]	1	Logs
➖	idle	memory utilization	+0.44	[+0.38, +0.49]	1	Logs
➖	tcp_dd_logs_filter_exclude	ingress throughput	+0.00	[-0.01, +0.01]	1	Logs
➖	uds_dogstatsd_to_api	ingress throughput	-0.00	[-0.00, +0.00]	1	Logs
➖	otel_to_otel_logs	ingress throughput	-0.05	[-0.87, +0.78]	1	Logs
➖	basic_py_check	% cpu utilization	-0.10	[-3.01, +2.80]	1	Logs
➖	tcp_syslog_to_blackhole	ingress throughput	-0.12	[-0.17, -0.07]	1	Logs
➖	file_tree	memory utilization	-0.25	[-0.36, -0.15]	1	Logs
➖	pycheck_lots_of_tags	% cpu utilization	-3.18	[-5.84, -0.53]	1	Logs

Bounds Checks

perf	experiment	bounds_check_name	replicates_passed
❌	idle	memory_usage	9/10

Explanation

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".

[dd-devflow]

🚂 MergeQueue: pull request added to the queue

The median merge time in main is 23m.

Use /merge -c to cancel this operation!