Repository navigation
Conversation
This comment has been minimized.
This comment has been minimized.
Contributor
🟢 Java Benchmark SLOs — All performance SLOs passed
PR vs. master results
Commit: Load and DaCapo benchmarks can be triggered manually in the GitLab pipeline. Results will appear in the Benchmarking Platform UI after completion. |
Kafka / producer-benchmarkParameters
See matching parameters
SummaryFound 0 performance improvements and 0 performance regressions! Performance is the same for 3 metrics, 0 unstable metrics. See unchanged results
|
Kafka / consumer-benchmarkParameters
See matching parameters
SummaryFound 0 performance improvements and 0 performance regressions! Performance is the same for 3 metrics, 0 unstable metrics. See unchanged results
|
This was referenced Nov 21, 2026
dougqh
force-pushed
the
dougqh/kafka-carrier-cache
branch
from
November 21, 2026 19:14
fe5dfd1 to
8131802
Compare
CompositePropagator.CarrierCache now implements a new IndexedCarrier (size/key/value by index) over a String[] sized for the handful of headers left after key filtering, growing on demand, instead of an ArrayList(32). HttpCodec.ExtractionCache reads an IndexedCarrier in place rather than copying every pair into a second ArrayList(32); other carriers are still copied. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
dougqh
force-pushed
the
dougqh/kafka-carrier-cache
branch
from
November 21, 2026 19:39
8131802 to
5bc558d
Compare
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What Does This Do
Extraction stops copying the carrier's headers twice. The composite propagator's cache becomes a small indexed array, and the tracing extractor reads it in place instead of copying every pair into a second list. Extracted contexts are unchanged.
Stacked on #13605.
Motivation
On a Kafka consumer, after #13600 removed
RecordHeader's copies, the two caches were the next-largest extraction allocation: anArrayList(32)each per record, about 3 GiB per profiled run (~12% of what the tracer still allocated), and their replay loops ~7–12% of the consumer thread's tracer time.Additional Notes
IndexedCarrier(new, components/context): read-onlysize()/key(i)/value(i)over pairs already in memory, valid only for the extraction it's handed to.CompositePropagator.CarrierCacheimplements it over aString[]sized for 8 pairs, growing on demand, instead of anArrayList(32); with key filtering a Kafka record keeps a handful of headers.HttpCodec.ExtractionCachereads anIndexedCarrierin place (computing the Datadog span id hex by one scan, as before); any other carrier is still copied.CarrierCacheTest(indexed reads, growth, visiting order).CompoundExtractionGoldenTestextracts through the composite propagator, so it covers the in-place read against its golden file.ExtractionCache.<init>gone (was 1.39 GiB per run) andCarrierCache.<init>down from 1.60 to 1.03 GiB; tracer-attributed allocation per run 24.2 → 22.0 GiB (−9%).🤖 Generated with Claude Code