tag aggregation smp tests by rayz · Pull Request #1252 · DataDog/saluki

rayz · 2026-03-20T15:10:39Z

Summary

Change Type

Bug fix
New feature
Non-functional (chore, refactoring, docs)
Performance

How did you test this PR?

References

Copilot

Pull request overview

This PR adds new SMP regression experiments for ADP to exercise tag-filterlist/tag-aggregation scenarios at scale, and adds transform-level telemetry for tag filtering outcomes.

Changes:

Add tagfilter_* SMP experiments (0 tags, 100 metrics, 500 metrics, and a 500-metrics no-traffic variant) and supporting shared filterlist fixtures.
Update experiment generation merge behavior to allow explicitly overriding lading generators with an empty list.
Add telemetry counters to the tag_filterlist transform and a unit test validating counter behavior.

Reviewed changes

Copilot reviewed 23 out of 23 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
test/smp/regression/adp/shared/tagfilter_500metrics.yaml	Adds a large shared `metric_tag_filterlist` fixture for 500-metric scenarios.
test/smp/regression/adp/shared/tagfilter_100metrics.yaml	Adds a shared `metric_tag_filterlist` fixture for 100-metric scenarios.
test/smp/regression/adp/generate_experiments.py	Allows an explicit empty overlay generator list to replace generators (no-traffic experiments).
test/smp/regression/adp/experiments.yaml	Introduces `tagfilter_base` template and new `tagfilter_*` experiments wiring shared fixtures.
test/smp/regression/adp/config.yaml	Updates lading version pin.
test/smp/regression/adp/cases/tagfilter_/*	Adds generated case directories for the new tagfilter experiments.
lib/saluki-components/src/transforms/tag_filterlist/telemetry.rs	Adds a telemetry helper to record hit/miss/noop/modified outcomes and removed-tag counts.
lib/saluki-components/src/transforms/tag_filterlist/mod.rs	Wires telemetry into the transform, returns filter outcomes, and adds a unit test for telemetry.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

test/smp/regression/adp/cases/tagfilter_500metrics/lading/lading.yaml

test/smp/regression/adp/cases/tagfilter_100metrics/lading/lading.yaml

test/smp/regression/adp/cases/tagfilter_0tags/lading/lading.yaml

Copilot · 2026-03-20T15:16:10Z

lib/saluki-components/src/transforms/tag_filterlist/mod.rs

        ComponentContext,
    },
    data_model::event::{metric::Metric, EventType},
+    observability::ComponentMetricsExt,


saluki-components has #![deny(warnings)], and ComponentMetricsExt is imported here but not used anywhere in the module. This will fail the build due to an unused import warning. Remove the import, or use the extension trait explicitly if it’s required for metrics wiring.

Suggested change

observability::ComponentMetricsExt,

pr-commenter · 2026-03-20T15:20:34Z

Binary Size Analysis (Agent Data Plane)

Target: 1d50af4 (baseline) vs 46578c9 (comparison) diff
Analysis Type: Stripped binaries (debug symbols excluded)
Baseline Size: 26.19 MiB
Comparison Size: 26.28 MiB
Size Change: +90.61 KiB (+0.34%)
Pass/Fail Threshold: +5%
Result: PASSED ✅

Changes by Module

Module	File Size	Symbols
`core`	+29.55 KiB	72
`saluki_components::transforms::tag_filterlist`	+27.56 KiB	18
`serde_core`	+8.78 KiB	8
`agent_data_plane::cli::run`	+6.86 KiB	2
`[sections]`	+6.08 KiB	6
`hashbrown`	+5.59 KiB	9
`saluki_config::dynamic::watcher`	-2.34 KiB	1
`saluki_context::hash::hash_context`	+2.15 KiB	1
`tokio`	+1.81 KiB	2
`[Unmapped]`	+1.52 KiB	1
`saluki_core::topology::blueprint`	+1.19 KiB	1
`saluki_context::context::Context`	+887 B	1
`smallvec`	+753 B	1
`saluki_context::resolver::TagsResolver`	-574 B	1
`saluki_components::encoders::datadog`	+379 B	2
`serde_json`	+281 B	1
`unicode_segmentation`	+104 B	1
`saluki_config::secrets::resolver`	-88 B	1
`figment`	+76 B	17
`saluki_components::destinations::prometheus`	+59 B	1

Detailed Symbol Changes

    FILE SIZE        VM SIZE    
 --------------  -------------- 
  +2.8% +43.7Ki  +2.9% +32.5Ki    [523 Others]
  [NEW] +12.8Ki  [NEW] +12.6Ki    _<saluki_components::transforms::tag_filterlist::TagFilterlist as saluki_core::components::transforms::Transform>::run::_{{closure}}::h963c4e73613d8740
  [NEW] +12.6Ki  [NEW] +12.4Ki    _<core::marker::PhantomData<T> as serde_core::de::DeserializeSeed>::deserialize::h823409a611a5cfb1
  [NEW] +10.7Ki  [NEW] +10.6Ki    <saluki_core::data_model::event::Event as core::clone::Clone>::clone.10049
  +6.0% +6.86Ki  +6.0% +6.86Ki    agent_data_plane::cli::run::create_topology::_{{closure}}::h4a1b0b37b7e2d03e
  [NEW] +5.66Ki  [NEW] +5.49Ki    serde_core::de::impls::_<impl serde_core::de::Deserialize for alloc::vec::Vec<T>>::deserialize::h34dcebfa9500250f
  [NEW] +5.52Ki  [NEW] +5.38Ki    <hickory_proto::rr::record_data::RData as core::clone::Clone>::clone.8502
  [NEW] +5.13Ki  [NEW] +5.00Ki    <hickory_proto::rr::record_data::RData as core::clone::Clone>::clone.11186
  [NEW] +4.76Ki  [NEW] +4.61Ki    _<core::marker::PhantomData<T> as serde_core::de::DeserializeSeed>::deserialize::h4e9adec8ddd65cfa
  [NEW] +4.71Ki  [NEW] +4.59Ki    <rustls::error::Error as core::clone::Clone>::clone.10025
  [NEW] +4.71Ki  [NEW] +4.59Ki    <rustls::error::Error as core::clone::Clone>::clone.11201
  [NEW] +4.26Ki  [NEW] +4.15Ki    saluki_components::transforms::tag_filterlist::compile_filters::hd808d986c0d7a078
  [NEW] +3.83Ki  [NEW] +3.71Ki    <webpki::error::Error as core::fmt::Debug>::fmt.11379
  [NEW] +3.66Ki  [NEW] +3.54Ki    <rustls::error::Error as core::fmt::Debug>::fmt.9647
  [DEL] -3.66Ki  [DEL] -3.54Ki    <rustls::error::Error as core::fmt::Debug>::fmt.9641
  [DEL] -3.83Ki  [DEL] -3.71Ki    <webpki::error::Error as core::fmt::Debug>::fmt.11359
  [DEL] -4.71Ki  [DEL] -4.59Ki    <rustls::error::Error as core::clone::Clone>::clone.11181
  [DEL] -4.71Ki  [DEL] -4.59Ki    <rustls::error::Error as core::clone::Clone>::clone.10017
  [DEL] -5.13Ki  [DEL] -5.00Ki    <hickory_proto::rr::record_data::RData as core::clone::Clone>::clone.11166
  [DEL] -5.52Ki  [DEL] -5.38Ki    <hickory_proto::rr::record_data::RData as core::clone::Clone>::clone.8496
  [DEL] -10.7Ki  [DEL] -10.6Ki    <saluki_core::data_model::event::Event as core::clone::Clone>::clone.10041
  +0.3% +90.6Ki  +0.3% +78.6Ki    TOTAL

pr-commenter · 2026-03-20T15:51:58Z

Regression Detector (Agent Data Plane)

This comment was omitted because it was over 65,536 characters.Please check the Gitlab Job logs to see its output.

webern

Caveat: I'm new here!

Mostly questions due to lack of exposure to the codebase. I ran out of time to try to run anything, so this was just a visual review.

webern · 2026-03-23T17:15:01Z

lib/saluki-components/src/transforms/tag_filterlist/telemetry.rs

+
+use super::FilterMetricTagsOutcome;
+
+#[derive(Clone)]


I know binary size is a big thing for us. Do we not derive Debug by default for that reason?

webern · 2026-03-23T17:32:56Z

lib/saluki-components/src/transforms/tag_filterlist/telemetry.rs

+use super::FilterMetricTagsOutcome;
+
+#[derive(Clone)]
+pub struct Telemetry {


I think structures like this should have doc comments, especially when pub. Or... to kind of say something similar but differently, Telemetry is a fairly overloaded name in a repository like this. Is there a pattern of having many Telemetry structs in the codebase for different purposes?

Perhaps TagFilterTelemetry?

Edit: ok I answered the question with Claude. Yes there is a pattern of telemetry.rs files and structs named telemetry.

My next question is whether it needs to be pub...

webern · 2026-03-23T17:42:45Z

test/smp/regression/adp/cases/tagfilter_0tags/experiment.yaml

+    DD_PROFILING_INLINED_FUNCTIONS: "true"
+checks:
+- name: memory_usage
+  description: Memory usage quality gate. This puts a bound on the total ADP memory usage.


Is this intended to describe the entire experiment, or just the check? Without looking at other experiment.yaml files (yet), it feels like what is missing is a story about why the experiment was added. For example we have these large metric names that are the same until the Nth character position and these are being added to verify the performance of a new tag filtering mechanism. Do we want some of that context and thinking to be added to either the lading.yaml or experiment.yaml files?

webern · 2026-03-23T17:47:06Z

test/smp/regression/adp/config.yaml

@@ -1,5 +1,5 @@
 lading:
-  version: 0.31.2


Suspicious, but I don't know what this is doing as of yet.

We're using a specific version of lading (the load generator for benchmarking the tag filtering behavior) that hasn't been released yet. edit: hmm this doesn't seem to be working 🤔

Copilot

Pull request overview

Copilot reviewed 23 out of 23 changed files in this pull request and generated no new comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot

Pull request overview

Copilot reviewed 23 out of 23 changed files in this pull request and generated no new comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

rayz · 2026-04-01T17:46:32Z

moved to 1296

tag aggregation smp tests

ea2a594

Copilot AI review requested due to automatic review settings March 20, 2026 15:10

dd-octo-sts bot added area/components Sources, transforms, and destinations. area/test All things testing: unit/integration, correctness, SMP regression, etc. labels Mar 20, 2026

Copilot started reviewing on behalf of rayz March 20, 2026 15:11 View session

Copilot AI reviewed Mar 20, 2026

View reviewed changes

webern approved these changes Mar 23, 2026

View reviewed changes

rayz added 2 commits March 24, 2026 15:14

Update to use master benchmarks

bcb525a

ci

4af6450

Copilot AI review requested due to automatic review settings March 24, 2026 21:04

Copilot started reviewing on behalf of rayz March 24, 2026 21:05 View session

Copilot AI reviewed Mar 24, 2026

View reviewed changes

rayz added 2 commits March 24, 2026 17:34

try this version of lading

17889ed

Update lading

235283c

Copilot AI review requested due to automatic review settings March 25, 2026 15:25

Copilot started reviewing on behalf of rayz March 25, 2026 15:26 View session

Copilot AI reviewed Mar 25, 2026

View reviewed changes

Trigger ci

46578c9

rayz closed this Apr 1, 2026

Conversation

rayz commented Mar 20, 2026

Summary

Change Type

How did you test this PR?

References

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI Mar 20, 2026

Choose a reason for hiding this comment

Uh oh!

pr-commenter bot commented Mar 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Binary Size Analysis (Agent Data Plane)

Changes by Module

Detailed Symbol Changes

Uh oh!

pr-commenter bot commented Mar 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Regression Detector (Agent Data Plane)

Uh oh!

webern left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

webern Mar 23, 2026

Choose a reason for hiding this comment

Uh oh!

webern Mar 23, 2026

Choose a reason for hiding this comment

Uh oh!

webern Mar 23, 2026

Choose a reason for hiding this comment

Uh oh!

webern Mar 23, 2026

Choose a reason for hiding this comment

Uh oh!

rayz Mar 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

rayz commented Apr 1, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

pr-commenter bot commented Mar 20, 2026 •

edited

Loading

pr-commenter bot commented Mar 20, 2026 •

edited

Loading

webern left a comment •

edited

Loading

rayz Mar 24, 2026 •

edited

Loading