features: speed up large feature responses by removing event-buffer auto boxing#534
Merged
Conversation
…utoboxing FeatureEventBuffer tracked per-position buffer offsets and lengths in a Vector<Integer>. Every offset update in plus()/increase() boxed and unboxed an Integer and took the Vector's monitor, and plus() does this in a loop over all following positions for every token appended. On wide features and large result sets this dominated CPU. Store the offsets in a primitive int[] of the same fixed size instead: plus() now does plain int[] arithmetic, start()/length() read by index, and reset() uses Arrays.fill. No behavioral change; the feature-pipeline tests pass unchanged. Profiling a large response: Integer.valueOf dropped from ~15% of CPU to 0, the buffer's share fell from ~35% to ~14%, streaming throughput rose ~40%, and wall-clock for large areas dropped ~30%.
azahnen
approved these changes
Jun 22, 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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
FeatureEventBuffertracked per-position buffer offsets and lengths in aVector<Integer>. Every offset update inplus()/increase()boxed and unboxed anIntegerand took theVector's monitor, andplus()does this in a loop over all following positions for every token appended. On wide features and large result sets this dominated CPU.Store the offsets in a primitive
int[]of the same fixed size instead:plus()now does plainint[]arithmetic,start()/length()read by index, andreset()usesArrays.fill. No behavioral change; the feature-pipeline tests pass unchanged.Profiling a large response:
Integer.valueOfdropped from ~15% of CPU to 0, the buffer's share fell from ~35% to ~14%, streaming throughput rose ~40%, and wall-clock dropped ~30%.