Conversation
Polar Signals Profiling ResultsLatest Run
Previous Runs (5)
Powered by Polar Signals Cloud |
Benchmarks: PolarSignals ProfilingVortex (geomean): 1.015x ➖ datafusion / vortex-file-compressed (1.015x ➖, 0↑ 1↓)
|
File Sizes: PolarSignals ProfilingFile Size Changes (1 files changed, +0.0% overall, 1↑ 0↓)
Totals:
|
Benchmarks: TPC-H SF=1 on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.968x ➖, 0↑ 0↓)
datafusion / vortex-compact (0.995x ➖, 0↑ 0↓)
datafusion / parquet (0.968x ➖, 2↑ 0↓)
datafusion / arrow (0.922x ➖, 6↑ 0↓)
duckdb / vortex-file-compressed (0.967x ➖, 1↑ 0↓)
duckdb / vortex-compact (0.952x ➖, 1↑ 0↓)
duckdb / parquet (0.976x ➖, 1↑ 0↓)
duckdb / duckdb (0.953x ➖, 3↑ 0↓)
Full attributed analysis
|
File Sizes: TPC-H SF=1 on NVMEFile Size Changes (8 files changed, -0.0% overall, 0↑ 8↓)
Totals:
|
Benchmarks: FineWeb NVMeVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (1.064x ➖, 0↑ 2↓)
datafusion / vortex-compact (1.052x ➖, 0↑ 1↓)
datafusion / parquet (1.084x ➖, 0↑ 2↓)
duckdb / vortex-file-compressed (1.061x ➖, 0↑ 1↓)
duckdb / vortex-compact (1.094x ➖, 0↑ 3↓)
duckdb / parquet (1.069x ➖, 0↑ 2↓)
Full attributed analysis
|
File Sizes: FineWeb NVMeFile Size Changes (1 files changed, +0.0% overall, 1↑ 0↓)
Totals:
|
Benchmarks: TPC-DS SF=1 on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.994x ➖, 0↑ 0↓)
datafusion / vortex-compact (0.996x ➖, 1↑ 1↓)
datafusion / parquet (0.991x ➖, 2↑ 1↓)
duckdb / vortex-file-compressed (1.008x ➖, 0↑ 2↓)
duckdb / vortex-compact (1.004x ➖, 1↑ 2↓)
duckdb / parquet (1.002x ➖, 0↑ 1↓)
duckdb / duckdb (0.996x ➖, 0↑ 1↓)
Full attributed analysis
|
File Sizes: TPC-DS SF=1 on NVMEFile Size Changes (24 files changed, +0.0% overall, 2↑ 22↓)
Totals:
|
Benchmarks: FineWeb S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.978x ➖, 0↑ 0↓)
datafusion / vortex-compact (0.901x ➖, 1↑ 0↓)
datafusion / parquet (0.863x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.053x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.933x ➖, 0↑ 0↓)
duckdb / parquet (1.007x ➖, 0↑ 0↓)
Full attributed analysis
|
Benchmarks: TPC-H SF=10 on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.943x ➖, 2↑ 0↓)
datafusion / vortex-compact (0.972x ➖, 0↑ 0↓)
datafusion / parquet (0.951x ➖, 0↑ 0↓)
datafusion / arrow (0.967x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.003x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.970x ➖, 0↑ 0↓)
duckdb / parquet (0.981x ➖, 0↑ 0↓)
duckdb / duckdb (0.977x ➖, 0↑ 0↓)
Full attributed analysis
|
File Sizes: TPC-H SF=10 on NVMEFile Size Changes (29 files changed, -0.0% overall, 0↑ 29↓)
Totals:
|
Benchmarks: Statistical and Population GeneticsVerdict: No clear signal (low confidence) duckdb / vortex-file-compressed (0.822x ✅, 11↑ 0↓)
duckdb / vortex-compact (0.838x ✅, 10↑ 0↓)
duckdb / parquet (0.878x ✅, 9↑ 0↓)
Full attributed analysis
|
File Sizes: Statistical and Population GeneticsFile Size Changes (2 files changed, +0.0% overall, 1↑ 1↓)
Totals:
|
Benchmarks: Random AccessVortex (geomean): 0.964x ➖ unknown / unknown (0.972x ➖, 1↑ 0↓)
|
Benchmarks: Clickbench on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.969x ➖, 1↑ 1↓)
datafusion / parquet (0.961x ➖, 2↑ 0↓)
duckdb / vortex-file-compressed (0.962x ➖, 2↑ 1↓)
duckdb / parquet (0.980x ➖, 1↑ 0↓)
duckdb / duckdb (0.945x ➖, 4↑ 0↓)
Full attributed analysis
|
File Sizes: Clickbench on NVMEFile Size Changes (110 files changed, +0.0% overall, 98↑ 12↓)
Totals:
|
Benchmarks: TPC-H SF=1 on S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.883x ➖, 4↑ 2↓)
datafusion / vortex-compact (0.930x ➖, 0↑ 2↓)
datafusion / parquet (0.932x ➖, 2↑ 0↓)
duckdb / vortex-file-compressed (0.946x ➖, 1↑ 0↓)
duckdb / vortex-compact (0.961x ➖, 0↑ 0↓)
duckdb / parquet (0.965x ➖, 0↑ 0↓)
Full attributed analysis
|
Benchmarks: CompressionVortex (geomean): 0.997x ➖ unknown / unknown (0.997x ➖, 0↑ 2↓)
|
Benchmarks: TPC-H SF=10 on S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (1.002x ➖, 0↑ 1↓)
datafusion / vortex-compact (1.000x ➖, 0↑ 0↓)
datafusion / parquet (0.938x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.009x ➖, 0↑ 0↓)
duckdb / vortex-compact (1.047x ➖, 0↑ 0↓)
duckdb / parquet (1.045x ➖, 0↑ 0↓)
Full attributed analysis
|
|
so compress time definitely improved, but there are regressions in file size. I think I know why (chooses FSST instead of Dict), and I can fix that |
d9a6ec5 to
9823b45
Compare
Signed-off-by: Connor Tsui <connor.tsui20@gmail.com>
Signed-off-by: Connor Tsui <connor.tsui20@gmail.com>
Signed-off-by: Connor Tsui <connor.tsui20@gmail.com>
Signed-off-by: Connor Tsui <connor.tsui20@gmail.com>
0a0058d to
94bcb3d
Compare
|
so it is quite hard to do this (or at least codex has not been able to find holistically good improvements, so I'm going to close this. |
Summary
Tracking issue: #7216
API Changes
TODO maybe?
Testing
Benchmarks run.