WordCache: FlatCache tricks - #2247
Merged
Merged
Conversation
Replaces the 4-way set-associative table with an open-addressed one that makes room by overwriting rather than by rebuilding. Two things drove it, both measured in cache_bench on 3 models x 18 corpora: the 4-way table locks ~0.4% of words out on bucket conflicts that open addressing simply does not have, and the alternative that fixes that (PR #2234's FlatCache) can only reclaim arena space by compacting every live entry into a second set of buffers -- so it carries two of everything, 12.6 MB per instance against this one's 2.7 MB. Nothing here needs a second copy: - Eviction overwrites a slot in place. Slots go empty -> occupied and never back, so a probe walk is never cut short by a hole and there are no tombstones and no backward shift. The cost is that the walk needs a bound, hence WINDOW. - Slots are self-contained at 32 B, two to a cache line. A <= 15-byte word packs into the u128 key, so a hit is a register compare with no memcmp and no key arena; up to 3 ids ride in the slot, so most hits never touch the id arena either. - What still needs an arena takes a run from a free list per exact length (MAX_LENGTH bounds every run, so there is no rounding waste), and gives it back on eviction. The arenas therefore hold the live set instead of every insert, and never need compacting. Retention is FlatCache's idea made local: a full window evicts its coldest slot and halves the frequencies it passed, so an entry has to keep being used to keep its place. Measured against the 4-way it replaces, at the shipped 65,536 capacity: 0.000% steady-state miss rate against 0.413%, and x1.12 median warm replay across the 54 model x corpus pairs. Against FlatCache it is a near-tie on speed (x0.97) for 22% of the memory, and x11.6 ahead on the one corpus where FlatCache's arena budget forces it to cull. `get` now takes &mut self: it bumps the entry's frequency and unpacks inline ids into a scratch field to hand back a slice. The one call site already held a &mut binding, so it needed no change.
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
SBrandeis
commented
Jul 27, 2026
Co-authored-by: Simon Brandeis <33657802+SBrandeis@users.noreply.github.com>
SBrandeis
changed the title
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.
PipelineTokenizer benchmark
9 / 10 models supported — PipelineTokenizer vs
tokenizersv0.23.1 (latest release) · ~10 kB inputs · add_special_tokens on · single thread + 1/2/4/8/max-thread sweep2f16acfce · 2026-07-28 22:06 UTC· Intel(R) Xeon(R) Platinum 8375C CPU @ 2.90GHz · 8 coresvs base branch (
709ef7af5) — per-model geomean ×speedup of this PR's PipelineTokenizer against the base branch's; regressions in red.Decode
Round-trip: v0.23.1
encode_fastproduces the id streams (same fixtures,add_special_tokens=true); both implementations decode those SAME ids withskip_special_tokens=false. MB/s counts decoded text bytes.bert-base-uncased — normalizer-heavy WordPiece · ×4.89 vs v0.23.1 · ×0.99 vs base · decode pending
Memory (RSS MB, load+encode): v0.23.1 12+0 (peak 12) · Pipeline 8+0 (peak 17)
deepseek-v4 — deepseek 3-regex split-heavy byte-level BPE · ×25.36 vs v0.23.1 · ×6.00 vs base · decode pending
Memory (RSS MB, load+encode): v0.23.1 62+0 (peak 68) · Pipeline 82+0 (peak 82)
Pre-tokenize:
classify + fsmvs regex engines — ns/byte, lower better. The fsm is the scalar jump-table in both pipe columns; SIMD / scalar is the classify pass (regex pre-tokenizers have no SIMD fsm).×vs= engine ÷ our pipeline (SIMD / scalar classify);onig&pcre2(JIT) are C,fancyis pure-Rust fancy-regex,logosis a compile-time DFA lexer (approximate grammar; n/a for deepseek).gemma-4 — byte-fallback BPE, Metaspace-style split (gemma-4) · ×2.09 vs v0.23.1 · ×1.09 vs base · decode pending
Memory (RSS MB, load+encode): v0.23.1 304+0 (peak 371) · Pipeline 275+0 (peak 371)
gpt2 — gpt2 ByteLevel regex · ×30.08 vs v0.23.1 · ×3.66 vs base · decode pending
Memory (RSS MB, load+encode): v0.23.1 25+2 (peak 27) · Pipeline 27+0 (peak 27)
Pre-tokenize:
classify + fsmvs regex engines — ns/byte, lower better. The fsm is the scalar jump-table in both pipe columns; SIMD / scalar is the classify pass (regex pre-tokenizers have no SIMD fsm).×vs= engine ÷ our pipeline (SIMD / scalar classify);onig&pcre2(JIT) are C,fancyis pure-Rust fancy-regex,logosis a compile-time DFA lexer (approximate grammar; n/a for deepseek).gpt-oss — o200k-regex byte-level BPE (gpt-oss) · ×19.88 vs v0.23.1 · ×3.93 vs base · decode pending
Memory (RSS MB, load+encode): v0.23.1 241+0 (peak 315) · Pipeline 234+0 (peak 316)
Pre-tokenize:
classify + fsmvs regex engines — ns/byte, lower better. The fsm is the scalar jump-table in both pipe columns; SIMD / scalar is the classify pass (regex pre-tokenizers have no SIMD fsm).×vs= engine ÷ our pipeline (SIMD / scalar classify);onig&pcre2(JIT) are C,fancyis pure-Rust fancy-regex,logosis a compile-time DFA lexer (approximate grammar; n/a for deepseek).glm-5.2 — cl100k-variant regex byte-level BPE (glm-5.2) · ×20.77 vs v0.23.1 · ×3.20 vs base · decode pending
Memory (RSS MB, load+encode): v0.23.1 169+0 (peak 231) · Pipeline 170+0 (peak 231)
Pre-tokenize:
classify + fsmvs regex engines — ns/byte, lower better. The fsm is the scalar jump-table in both pipe columns; SIMD / scalar is the classify pass (regex pre-tokenizers have no SIMD fsm).×vs= engine ÷ our pipeline (SIMD / scalar classify);onig&pcre2(JIT) are C,fancyis pure-Rust fancy-regex,logosis a compile-time DFA lexer (approximate grammar; n/a for deepseek).llama-2 — model-bounded BPE, no pre-tokenizer · ×4.05 vs v0.23.1 · ×1.13 vs base · decode pending
Memory (RSS MB, load+encode): v0.23.1 18+0 (peak 23) · Pipeline 23+0 (peak 23)
llama-3 — cl100k-regex byte-level BPE (llama-3), single regex · ×21.96 vs v0.23.1 · ×3.09 vs base · decode pending
Memory (RSS MB, load+encode): v0.23.1 73+0 (peak 95) · Pipeline 92+0 (peak 95)
Pre-tokenize:
classify + fsmvs regex engines — ns/byte, lower better. The fsm is the scalar jump-table in both pipe columns; SIMD / scalar is the classify pass (regex pre-tokenizers have no SIMD fsm).×vs= engine ÷ our pipeline (SIMD / scalar classify);onig&pcre2(JIT) are C,fancyis pure-Rust fancy-regex,logosis a compile-time DFA lexer (approximate grammar; n/a for deepseek).mistral-small-4 — tekken byte-level BPE, 1k added specials (mistral-small-4) · ×7.71 vs v0.23.1 · ×2.15 vs base · decode pending
Memory (RSS MB, load+encode): v0.23.1 152+0 (peak 194) · Pipeline 108+0 (peak 194)
Pre-tokenize:
classify + fsmvs regex engines — ns/byte, lower better. The fsm is the scalar jump-table in both pipe columns; SIMD / scalar is the classify pass (regex pre-tokenizers have no SIMD fsm).×vs= engine ÷ our pipeline (SIMD / scalar classify);onig&pcre2(JIT) are C,fancyis pure-Rust fancy-regex,logosis a compile-time DFA lexer (approximate grammar; n/a for deepseek).Not yet supported:
t5-base