Use handle-based tag storage in DogStatsD #1691
Merged
+176
−104
Conversation
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
Replace Vec<String> tag storage with Vec<Tag> containing Pool handles. This eliminates ~260K String allocations during DogStatsD initialization by storing tag references as (offset, length) handles into the string pool. Tags are formatted to "key:value" strings at serialization time via Display impls rather than during construction. With default configs (~7500 contexts × ~26 tags), this significantly reduces init-time memory allocation overhead. Benchmark results (payloadtool + hyperfine): - Main: 175.2ms ± 5.8ms - This: 118.6ms ± 2.6ms - Improvement: ~32% faster
blt
force-pushed
the
blt/optimize-dogstatsd-handle-tags
branch
from
70a7929 to
c9eb5b0
Compare
GeorgeHahn
reviewed
Jan 12, 2026
| // Use handle-based tagset from core module directly, avoiding String | ||
| // allocations. Formatting happens at serialization time with O(1) handle | ||
| // lookups. | ||
| pub(crate) type Tagset = crate::common::tags::Tagset; |
Contributor
There was a problem hiding this comment.
Light rec to use crate::...::tagset here and move the comment onto that type if relevant.
GeorgeHahn
approved these changes
Jan 12, 2026
blt
added a commit
that referenced
this pull request
Jan 15, 2026
Refactor skills to reduce token usage while preserving functionality: - Extract shared BENCHMARKING.md and DB_SCHEMA.md references - Remove duplicated bash blocks (worktree, criterion, hyperfine) - Convert verbose tutorials to concise checklists - Trust LLM knowledge of bash/Rust basics Based on analysis of merged PRs, establishes a standard commit format: - Title: Action verb + target + technique (no prefixes) - Body: Brief explanation + benchmark results - ONE commit per optimization (includes db.yaml update) Examples from merged PRs analyzed: - Reuse buffer in Apache Common serialization (#1690) - Eliminate allocations in Splunk HEC serialization (#1694) - Use handle-based tag storage in DogStatsD (#1691) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> Signed-off-by: Brian L. Troutwine <brian.troutwine@datadoghq.com>
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.
Summary
Vec<String>tag storage withVec<Tag>containing Pool handlesPerformance
With default configs (~7500 contexts × ~26 tags), this significantly reduces init-time memory allocation overhead.
Benchmark results (payloadtool + hyperfine):