[9.5](backport #51714) otel(tests): align BenchmarkFilebeatOTelThroughputMockES with presets#51898
Merged
Conversation
…#51714) * otel(tests): align BenchmarkFilebeatOTelThroughputMockES with presets The current benchmark options don't reflect what Elastic Agent actually renders into otel-merged.yaml for a real deployment. Turned the benchmark into a table test covering all presets, settings pulled from a 9.5.0 bb benchmark diagnostics bundles: preset mem queue (events/min_events/timeout) exporter batch (max/min/flush) conns/consumers/queue_size throughput 12800/1600/5s 1600/1600/5s 4/4/12800 balanced 3200/1600/10s 1600/1600/10s 1/1/3200 scale 3200/1600/20s 1600/1600/20s 1/1/3200 latency 4100/2050/1s 50/50/1s 1/1/4100 The total events processed by the benchmark are a multiple of the queue size, to avoid waiting on the flush timeout for the last batches which skews the benchmark numbers. Also expanded the benchmark to have 3 filebeatreceiver instances, each in its own pipeline, with a beatprocessor. Measured events/sec at different -cpu values (go test -bench=... -cpu=N). At the host's full core count, results are dominated by the ES exporter, batching, and I/O rather than processor CPU cost: preset -cpu=32 (full) -cpu=2 throughput ~97.6k ~34.8k balanced ~55.8k ~33.9k scale ~56.2k ~33.5k latency ~35.4k ~26.6k -cpu=2 approximates a throttled 4-core/cpu50 host, closer to the real benchmark suite's methodology, and is where processor-side CPU improvements are expected to actually show up in this benchmark's numbers. * reuse presets from es output (cherry picked from commit d6a567b)
botelastic
Bot
added
the
needs_team
github-actions
Bot
added
Team:Elastic-Agent-Data-Plane
Contributor
🤖 GitHub commentsJust comment with:
|
|
Pinging @elastic/elastic-agent-data-plane (Team:Elastic-Agent-Data-Plane) |
botelastic
Bot
removed
the
needs_team
mauri870
approved these changes
Jul 13, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Proposed commit message
The current benchmark options don't reflect what Elastic Agent actually renders into otel-merged.yaml for a real deployment.
Turned the benchmark into a table test covering all presets, settings pulled from a 9.5.0 bb benchmark diagnostics bundles (otel-merged.yaml):
ES output presets for reference https://www.elastic.co/docs/reference/beats/filebeat/elasticsearch-output#_preset.
The total events processed by the benchmark are a multiple of the queue size, to avoid waiting on the flush timeout for the last batches which skews the benchmark numbers.
Also expanded the benchmark to have 3 filebeatreceiver instances, each in its own pipeline, with a beatprocessor (add_fields + add_host_metadata).
Checklist
stresstest.shscript to run them under stress conditions and race detector to verify their stability../changelog/fragmentsusing the changelog tool.How to test this PR locally