Bump Datafusion to 55 - #9334
Conversation
Merging this PR will improve performance by 49.81%
|
c8b775d to
9dd15ac
Compare
Polar Signals Profiling ResultsLatest Run
Previous Runs (55)
Powered by Polar Signals Cloud |
Benchmarks: String Encoding 📖Commits: PR vortex / vortex-file-compressed / ms (1.005x ➖, 0↑ 0↓)
vortex / vortex-file-compressed / % (1.000x ➖, 0↑ 0↓)
|
🚨🚨🚨❌❌❌ SQL BENCHMARK FAILED ❌❌❌🚨🚨🚨Benchmark |
🚨🚨🚨❌❌❌ SQL BENCHMARK FAILED ❌❌❌🚨🚨🚨Benchmark |
Benchmarks: PolarSignals Profiling 📖Commits: PR datafusion / vortex-file-compressed / ns (0.994x ➖, 2↑ 1↓)
No file size changes detected. |
🚨🚨🚨❌❌❌ SQL BENCHMARK FAILED ❌❌❌🚨🚨🚨Benchmark |
Benchmarks: FineWeb NVMe 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (1.001x ➖, 0↑ 0↓)
datafusion / vortex-compact / ns (1.000x ➖, 0↑ 0↓)
datafusion / parquet / ns (0.990x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed / ns (0.936x ➖, 1↑ 0↓)
duckdb / vortex-compact / ns (1.031x ➖, 0↑ 1↓)
duckdb / parquet / ns (0.996x ➖, 0↑ 0↓)
No file size changes detected. |
🚨🚨🚨❌❌❌ SQL BENCHMARK FAILED ❌❌❌🚨🚨🚨Benchmark |
🚨🚨🚨❌❌❌ SQL BENCHMARK FAILED ❌❌❌🚨🚨🚨Benchmark |
Benchmarks: Clickbench Sorted on NVME 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (1.028x ➖, 0↑ 1↓)
datafusion / vortex-compact / ns (1.018x ➖, 0↑ 1↓)
datafusion / parquet / ns (0.835x ✅, 3↑ 0↓)
duckdb / vortex-file-compressed / ns (1.007x ➖, 0↑ 1↓)
duckdb / vortex-compact / ns (0.985x ➖, 1↑ 0↓)
duckdb / parquet / ns (1.003x ➖, 0↑ 1↓)
File Size Changes (200 files changed, +0.0% overall, 105↑ 95↓)
Totals:
|
Benchmarks: Appian on NVME 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-compact / ns (0.953x ➖, 1↑ 0↓)
datafusion / parquet / ns (0.892x ✅, 5↑ 0↓)
duckdb / vortex-compact / ns (1.008x ➖, 0↑ 0↓)
duckdb / parquet / ns (0.998x ➖, 0↑ 0↓)
File Size Changes (10 files changed, -63.8% overall, 0↑ 10↓)
Totals:
|
Benchmarks: FineWeb S3 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.943x ➖, 1↑ 1↓)
datafusion / vortex-compact / ns (0.952x ➖, 0↑ 0↓)
datafusion / parquet / ns (1.028x ➖, 0↑ 1↓)
duckdb / vortex-file-compressed / ns (0.955x ➖, 0↑ 0↓)
duckdb / vortex-compact / ns (1.133x ➖, 1↑ 1↓)
duckdb / parquet / ns (0.842x ➖, 1↑ 0↓)
|
Benchmarks: Statistical and Population Genetics 📖Commits: PR How to read Verdict and Engines
duckdb / vortex-file-compressed / ns (0.992x ➖, 2↑ 2↓)
duckdb / vortex-compact / ns (0.997x ➖, 2↑ 2↓)
duckdb / parquet / ns (0.998x ➖, 0↑ 0↓)
No file size changes detected. |
Benchmarks: Clickbench on NVME 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.996x ➖, 0↑ 0↓)
datafusion / vortex-compact / ns (1.004x ➖, 0↑ 1↓)
datafusion / parquet / ns (0.979x ➖, 4↑ 0↓)
duckdb / vortex-file-compressed / ns (0.994x ➖, 1↑ 1↓)
duckdb / vortex-compact / ns (0.995x ➖, 1↑ 0↓)
duckdb / parquet / ns (0.995x ➖, 2↑ 1↓)
No file size changes detected. |
Benchmarks: Random Access 📖Commits: PR How to read Verdict and Engines
random-access / vortex-file-compressed / ns (0.928x ➖, 1↑ 0↓)
random-access / parquet / ns (2.531x ❌, 0↑ 8↓)
random-access / lance / ns (1.249x ❌, 0↑ 1↓)
|
Benchmarks: Compression 📖Commits: PR vortex / vortex-file-compressed / ns (0.966x ➖, 2↑ 0↓)
vortex / vortex-file-compressed / bytes (1.000x ➖, 0↑ 0↓)
vortex / vortex-file-compressed / ratio (1.008x ➖, 1↑ 6↓)
vortex / parquet / ns (0.954x ➖, 6↑ 0↓)
vortex / parquet / bytes (1.000x ➖, 0↑ 0↓)
|
7152f4c to
27251c4
Compare
Benchmarks: Statistical and Population Genetics 📖Commits: PR How to read Verdict and Engines
duckdb / vortex-file-compressed / ns (1.023x ➖, 1↑ 3↓)
duckdb / parquet / ns (0.993x ➖, 0↑ 0↓)
No file size changes detected. |
Benchmarks: FineWeb S3 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (1.061x ➖, 0↑ 0↓)
datafusion / parquet / ns (0.896x ➖, 1↑ 0↓)
duckdb / vortex-file-compressed / ns (0.828x ➖, 1↑ 1↓)
duckdb / parquet / ns (0.948x ➖, 2↑ 1↓)
|
Benchmarks: Clickbench on NVME 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.998x ➖, 0↑ 0↓)
datafusion / parquet / ns (0.986x ➖, 2↑ 1↓)
duckdb / vortex-file-compressed / ns (1.008x ➖, 1↑ 1↓)
duckdb / parquet / ns (1.010x ➖, 1↑ 1↓)
No file size changes detected. |
Benchmarks: TPC-H SF=1 on S3 📖Commits: PR How to read Verdict and Engines
datafusion / vortex-file-compressed / ns (0.955x ➖, 1↑ 0↓)
datafusion / parquet / ns (0.892x ➖, 2↑ 0↓)
duckdb / vortex-file-compressed / ns (0.941x ➖, 1↑ 0↓)
duckdb / parquet / ns (0.999x ➖, 0↑ 0↓)
|
|
@AdamGS you tried so hard in tpc-ds |
|
I know :( I have another PR to DF to address this - apache/datafusion#24371 |
|
Indeed apache/datafusion#24371 is the missing part to make it usable |
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
DataFusion 55 tightens the lower bounds of three crates that the workspace also declares directly: datafusion-common needs chrono 0.4.45, datafusion-functions needs memchr 2.8.1, and datafusion-macros needs syn 3.0.2. Our lower bounds stayed one patch behind, so `cargo minimal-versions check --direct` could not resolve a version that satisfied both sides and CI failed before it compiled anything. Cargo.lock already selects versions above the new floors, so this changes no resolved dependency for a normal build. Signed-off-by: Robert Kruszewski <robert@spiraldb.com>
Signed-off-by: Robert Kruszewski <github@robertk.io>
a65d744 to
325bfa3
Compare
Resolves the Cargo.toml conflict on the tpchgen pins. develop moved `tpchgen`/`tpchgen-arrow` from a git-pinned 2.0.2 to the released 3.0.0 as part of "Bump Datafusion to 55" (#9334); this branch still carried the pin and DataFusion 54. The pin and the release go together, so take develop's side: the merge brings DataFusion 55, which is the version 3.0.0 pairs with. The conflict region also duplicated the `vortex-array-macros` workspace dependency, which develop already lists in its sorted position among the crates published by this project. Dropped the stray copy; `cargo metadata` rejected the manifest until it was gone. Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Rationale for this change
Periodic update to DataFusion
What changes are included in this PR?
55.0.0(Jul / Aug 2026) apache/datafusion#22393)~ Currently this PR also includes bumping our MSRV to 1.95 because
geoarrowrequires that (DataFusion only requires 1.94) ~ Now #9334.What APIs are changed? Are there any user-facing changes?
Just the dependency changes