IN LIST: reinterpret FixedSizeBinary for primitive fast paths#23018
Draft
geoffreyclaude wants to merge 11 commits into
Draft
IN LIST: reinterpret FixedSizeBinary for primitive fast paths#23018geoffreyclaude wants to merge 11 commits into
geoffreyclaude wants to merge 11 commits into
Conversation
10 tasks
geoffreyclaude
force-pushed
the
perf/in_list_fixed_size_binary_filter
branch
3 times, most recently
from
70c420f to
098e0a6
Compare
geoffreyclaude
changed the title
Contributor
Author
|
run benchmark in_list_strategy |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing perf/in_list_fixed_size_binary_filter (098e0a6) to c7e9284 (merge-base) diff using: in_list_strategy File an issue against this benchmark runner |
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagein_list_strategy — base (merge-base)
in_list_strategy — branch
File an issue against this benchmark runner |
Contributor
Author
|
run benchmark in_list |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing perf/in_list_fixed_size_binary_filter (098e0a6) to c7e9284 (merge-base) diff using: in_list File an issue against this benchmark runner |
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagein_list — base (merge-base)
in_list — branch
File an issue against this benchmark runner |
geoffreyclaude
force-pushed
the
perf/in_list_fixed_size_binary_filter
branch
from
098e0a6 to
ce699c4
Compare
This was referenced Jun 18, 2026
geoffreyclaude
force-pushed
the
perf/in_list_fixed_size_binary_filter
branch
2 times, most recently
from
867992e to
2ac665f
Compare
Replaces HashSet<u8> with a 32-byte stack-allocated bitmap. Provides O(1) membership testing via bit-shifting, significantly reducing memory overhead and improving cache locality. Triggers for UInt8 arrays.
Implements an 8 KB heap-allocated bitmap for UInt16. Maintains O(1) performance while handling the larger value space. Triggers for UInt16 arrays.
Introduces zero-copy buffer reinterpretation to allow signed integers and other 1 or 2-byte primitive types (e.g. Float16) to use the high-performance bitmap filters. Triggers for all types with 1-byte or 2-byte width.
Adds a const-generic unrolled comparison chain that avoids CPU branching. Outperforms hash lookups for very small lists. Triggers for primitives when list size <= 32 (4-byte), 16 (8-byte), or 4 (16-byte).
Implements a fast hash table using open addressing with linear probing and a 25% load factor. Replaces the legacy HashSet for primitives, reducing indirection. Triggers for primitives when list size exceeds branchless thresholds.
Introduces a two-stage filter for ByteView types. Stage 1 uses a fast DirectProbeFilter on masked views (len + prefix) for quick rejection; Stage 2 performs full verification only for potential long-string matches. Triggers for Utf8View and BinaryView.
FixedSizeBinary(N) arrays share the same contiguous buffer layout as primitive arrays, so for power-of-2 widths (1, 2, 4, 8, 16) we can zero-copy reinterpret them and use the optimized primitive filters (bitmap, branchless, hash) instead of falling through to the NestedTypeFilter fallback.
geoffreyclaude
force-pushed
the
perf/in_list_fixed_size_binary_filter
branch
from
2ac665f to
b62beb4
Compare
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.
Which issue does this PR close?
INperformance with specialized implementations #19390.Rationale for this change
FixedSizeBinarymeans every value has the same number of bytes. For widths 1, 2, 4, 8, and 16, those bytes have the same shape as the primitive values optimized earlier in the stack.That lets DataFusion reuse the existing fast paths without copying the bytes:
For example, a
FixedSizeBinary(4)value is four bytes wide, just like aUInt32. The bytes can be checked by the same fixed-width lookup machinery. The value is still treated as binary data; this is only an internal lookup representation.Other fixed-size binary widths stay on the generic fallback path.
What changes are included in this PR?
FixedSizeBinary(1)andFixedSizeBinary(2)through the bitmap filters.FixedSizeBinary(4),(8), and(16)through branchless or direct-probe filters based on list size.FixedSizeBinaryneedles.FixedSizeBinarywidths onArrayStaticFilter.Are these changes tested?
Yes.
cargo fmt --all --checkcargo test -p datafusion-physical-expr fixed_size_binary --libcargo test -p datafusion-physical-expr test_in_list_from_array_type_combinations --libcargo test -p datafusion-physical-expr reinterpreted_ --libcargo test -p datafusion-physical-expr in_list_binary_types --libcargo clippy -p datafusion-physical-expr --all-targets --all-features -- -D warningsAre there any user-facing changes?
No. This is an internal performance optimization only.
Local benchmark snapshot
Benchmark command:
Method: compare adjacent saved baselines using raw Criterion sample minima (
min(time / iters)). Lower is better; changes within +/-5% are treated as noise.Compared baselines: #23016 -> #23018
Relevant scope: FixedSizeBinary rows.
Summary: 8 relevant rows, 8 faster, 0 slower, 0 within +/-5%.
fixed_size_binary/fsb16/list=10000/match=0%fixed_size_binary/fsb16/list=10000/match=50%fixed_size_binary/fsb16/list=256/match=0%fixed_size_binary/fsb16/list=256/match=50%fixed_size_binary/fsb16/list=4/match=0%fixed_size_binary/fsb16/list=4/match=50%fixed_size_binary/fsb16/list=64/match=0%fixed_size_binary/fsb16/list=64/match=50%