Lazily untranspose DeltaArray validity take #2#8890
Open
robert3005 wants to merge 3 commits into
Open
Conversation
The reapplied lazy Delta validity (#8725) broke in two ways once arrays went through slicing or a file round-trip: - Delta::validity required the deltas' storage validity to be a raw Bool array, but Validity::Array may be lazily encoded (e.g. a vortex.slice wrapper when the validity encoding has no static slice reduction, or a re-encoded array after a file round-trip), failing with "DeltaArray storage validity must be a BoolArray, got vortex.slice". TransposedBool now wraps the validity as a child array of any encoding and canonicalizes it at execution time, where an ExecutionCtx is available. - untranspose_bitbuffer asserted 8-byte alignment, panicking with "Transposed buffer must be 8 byte aligned" on buffers viewed at arbitrary byte offsets (e.g. file segments). The transpose routines now normalize sliced views to chunk-aligned bytes and fall back to a copy path with unaligned-safe loads, keeping the in-place fast path for uniquely-owned aligned buffers. This also fixes bit-offset views being untransposed from the wrong chunk boundaries. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Signed-off-by: Robert Kruszewski <github@robertk.io>
robert3005
added
action/benchmark
github-actions
Bot
removed
the
action/benchmark
Contributor
Author
|
This fixes the cases where the validity has been transformed. Need to investigate more how the misalignment happens |
Contributor
Polar Signals Profiling ResultsLatest Run
Previous Runs (1)
Powered by Polar Signals Cloud |
Contributor
Benchmarks: Vortex queries 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.077x ➖, 0↑ 1↓)
datafusion / parquet (1.027x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.067x ➖, 0↑ 0↓)
duckdb / parquet (1.011x ➖, 0↑ 0↓)
No file size changes detected. |
Contributor
Benchmarks: PolarSignals Profiling 📖Vortex (geomean): 1.000x ➖ How to read Verdict and Engines
datafusion / vortex-file-compressed (1.000x ➖, 1↑ 0↓)
No file size changes detected. |
Contributor
Benchmarks: FineWeb NVMe 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.809x ✅, 8↑ 0↓)
datafusion / vortex-compact (0.855x ✅, 9↑ 0↓)
datafusion / parquet (0.833x ✅, 9↑ 0↓)
duckdb / vortex-file-compressed (0.807x ✅, 9↑ 0↓)
duckdb / vortex-compact (0.847x ✅, 9↑ 0↓)
duckdb / parquet (0.882x ✅, 8↑ 0↓)
No file size changes detected. |
Contributor
Benchmarks: TPC-H SF=1 on NVME 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.997x ➖, 0↑ 1↓)
datafusion / vortex-compact (1.047x ➖, 1↑ 5↓)
datafusion / parquet (1.097x ➖, 0↑ 9↓)
datafusion / arrow (1.014x ➖, 0↑ 1↓)
duckdb / vortex-file-compressed (0.997x ➖, 0↑ 1↓)
duckdb / vortex-compact (1.028x ➖, 0↑ 1↓)
duckdb / parquet (0.978x ➖, 1↑ 0↓)
duckdb / duckdb (1.036x ➖, 0↑ 2↓)
No file size changes detected. |
Contributor
Benchmarks: TPC-DS SF=1 on NVME 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.120x ❌, 0↑ 63↓)
datafusion / vortex-compact (1.095x ➖, 0↑ 43↓)
datafusion / parquet (1.121x ❌, 0↑ 65↓)
duckdb / vortex-file-compressed (1.095x ➖, 0↑ 44↓)
duckdb / vortex-compact (1.014x ➖, 0↑ 2↓)
duckdb / parquet (1.058x ➖, 0↑ 10↓)
duckdb / duckdb (1.008x ➖, 1↑ 2↓)
No file size changes detected. |
Contributor
Benchmarks: TPC-H SF=1 on S3 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.931x ➖, 3↑ 1↓)
datafusion / vortex-compact (0.970x ➖, 0↑ 0↓)
datafusion / parquet (0.973x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.023x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.981x ➖, 0↑ 0↓)
duckdb / parquet (1.023x ➖, 0↑ 0↓)
|
Contributor
Benchmarks: FineWeb S3 📖Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.094x ➖, 0↑ 2↓)
datafusion / vortex-compact (0.906x ➖, 0↑ 0↓)
datafusion / parquet (1.051x ➖, 0↑ 1↓)
duckdb / vortex-file-compressed (0.981x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.983x ➖, 0↑ 0↓)
duckdb / parquet (1.000x ➖, 0↑ 0↓)
|
Contributor
Benchmarks: Statistical and Population Genetics 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
duckdb / vortex-file-compressed (1.041x ➖, 0↑ 1↓)
duckdb / vortex-compact (1.047x ➖, 0↑ 2↓)
duckdb / parquet (1.016x ➖, 0↑ 0↓)
No file size changes detected. |
Contributor
Benchmarks: Clickbench Sorted on NVME 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.037x ➖, 1↑ 2↓)
datafusion / parquet (0.975x ➖, 1↑ 0↓)
duckdb / vortex-file-compressed (0.892x ✅, 5↑ 0↓)
duckdb / parquet (0.882x ✅, 7↑ 0↓)
duckdb / duckdb (0.936x ➖, 1↑ 0↓)
File Size Changes (201 files changed, -0.0% overall, 96↑ 105↓)
Totals:
|
Contributor
Benchmarks: Clickbench on NVME 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.008x ➖, 0↑ 2↓)
datafusion / parquet (0.992x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.003x ➖, 1↑ 2↓)
duckdb / parquet (0.993x ➖, 0↑ 0↓)
duckdb / duckdb (1.003x ➖, 1↑ 0↓)
File Size Changes (1 files changed, -0.0% overall, 0↑ 1↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=10 on NVME 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.929x ➖, 0↑ 0↓)
datafusion / vortex-compact (0.958x ➖, 0↑ 0↓)
datafusion / parquet (0.898x ✅, 11↑ 0↓)
datafusion / arrow (0.857x ✅, 15↑ 0↓)
duckdb / vortex-file-compressed (1.000x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.999x ➖, 0↑ 0↓)
duckdb / parquet (0.997x ➖, 0↑ 0↓)
duckdb / duckdb (1.003x ➖, 0↑ 0↓)
No file size changes detected. |
Contributor
Benchmarks: Random Access 📖Vortex (geomean): 1.017x ➖ How to read Verdict and Engines
unknown / unknown (1.003x ➖, 2↑ 1↓)
|
Contributor
Benchmarks: Appian on NVME 📖Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.995x ➖, 0↑ 0↓)
datafusion / parquet (0.996x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.983x ➖, 0↑ 0↓)
duckdb / parquet (0.995x ➖, 0↑ 0↓)
duckdb / duckdb (0.973x ➖, 0↑ 0↓)
File Size Changes (1 files changed, -0.0% overall, 0↑ 1↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=10 on S3 📖Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.031x ➖, 1↑ 2↓)
datafusion / vortex-compact (1.011x ➖, 0↑ 0↓)
datafusion / parquet (1.012x ➖, 1↑ 1↓)
duckdb / vortex-file-compressed (0.998x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.983x ➖, 0↑ 0↓)
duckdb / parquet (1.005x ➖, 0↑ 0↓)
|
Contributor
Benchmarks: Compression 📖Vortex (geomean): 0.996x ➖ How to read Verdict and Engines
unknown / unknown (0.990x ➖, 3↑ 1↓)
|
joseph-isaacs
added
the
action/benchmark
github-actions
Bot
removed
the
action/benchmark
Contributor
|
we really should fix this s3 benchmarks |
Contributor
Author
|
I think #8903 will make it better |
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.
Lazily untranpose DeltaArray validity array