Conversation
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk> # Conflicts: # encodings/fastlanes/src/bitpacking/vtable/mod.rs # vortex-array/src/vtable/mod.rs
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Merging this PR will degrade performance by 23.78%
Performance Changes
Comparing Footnotes
|
Polar Signals Profiling ResultsLatest Run
Previous Runs (1)
Powered by Polar Signals Cloud |
Benchmarks: PolarSignals ProfilingVortex (geomean): 0.841x ✅ datafusion / vortex-file-compressed (0.841x ✅, 9↑ 0↓)
|
Benchmarks: TPC-H SF=1 on NVMEVerdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.733x ✅, 22↑ 0↓)
datafusion / vortex-compact (0.753x ✅, 22↑ 0↓)
datafusion / parquet (0.838x ✅, 18↑ 1↓)
datafusion / arrow (0.666x ✅, 22↑ 0↓)
duckdb / vortex-file-compressed (0.743x ✅, 22↑ 0↓)
duckdb / vortex-compact (0.792x ✅, 21↑ 0↓)
duckdb / parquet (0.896x ✅, 15↑ 3↓)
duckdb / duckdb (0.804x ✅, 21↑ 0↓)
Full attributed analysis
|
Benchmarks: FineWeb NVMeVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.991x ➖, 2↑ 1↓)
datafusion / vortex-compact (0.978x ➖, 0↑ 0↓)
datafusion / parquet (0.979x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.998x ➖, 1↑ 1↓)
duckdb / vortex-compact (1.009x ➖, 0↑ 1↓)
duckdb / parquet (0.984x ➖, 0↑ 0↓)
Full attributed analysis
|
encodings/fsst/src/array.rs
Outdated
| &array.codes_array | ||
| impl ValidityVTable<FSST> for FSST { | ||
| fn validity(array: &FSSTArray) -> VortexResult<Validity> { | ||
| Ok(array.codes().validity()) |
There was a problem hiding this comment.
you have the validity logic in codes that's a lot simpler than getting codes
Benchmarks: TPC-DS SF=1 on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.973x ➖, 2↑ 0↓)
datafusion / vortex-compact (1.017x ➖, 3↑ 4↓)
datafusion / parquet (0.995x ➖, 0↑ 2↓)
duckdb / vortex-file-compressed (0.938x ➖, 16↑ 0↓)
duckdb / vortex-compact (0.945x ➖, 12↑ 0↓)
duckdb / parquet (0.929x ➖, 29↑ 0↓)
duckdb / duckdb (1.038x ➖, 1↑ 10↓)
Full attributed analysis
|
Benchmarks: TPC-H SF=10 on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (1.077x ➖, 0↑ 7↓)
datafusion / vortex-compact (1.003x ➖, 2↑ 1↓)
datafusion / parquet (1.097x ➖, 0↑ 9↓)
datafusion / arrow (1.068x ➖, 1↑ 9↓)
duckdb / vortex-file-compressed (0.979x ➖, 4↑ 6↓)
duckdb / vortex-compact (0.944x ➖, 3↑ 0↓)
duckdb / parquet (1.002x ➖, 0↑ 0↓)
duckdb / duckdb (0.990x ➖, 1↑ 0↓)
Full attributed analysis
|
Benchmarks: TPC-H SF=1 on S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (1.289x ➖, 0↑ 11↓)
datafusion / vortex-compact (1.046x ➖, 3↑ 6↓)
datafusion / parquet (1.561x ❌, 0↑ 15↓)
duckdb / vortex-file-compressed (1.117x ➖, 0↑ 3↓)
duckdb / vortex-compact (0.947x ➖, 1↑ 0↓)
duckdb / parquet (1.068x ➖, 0↑ 3↓)
Full attributed analysis
|
Benchmarks: FineWeb S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.861x ➖, 3↑ 1↓)
datafusion / vortex-compact (1.016x ➖, 0↑ 0↓)
datafusion / parquet (1.135x ➖, 0↑ 1↓)
duckdb / vortex-file-compressed (1.019x ➖, 1↑ 2↓)
duckdb / vortex-compact (1.002x ➖, 1↑ 0↓)
duckdb / parquet (0.972x ➖, 0↑ 0↓)
Full attributed analysis
|
Benchmarks: Random AccessVortex (geomean): 0.913x ➖ unknown / unknown (0.950x ➖, 14↑ 4↓)
|
Benchmarks: Statistical and Population GeneticsVerdict: No clear signal (low confidence) duckdb / vortex-file-compressed (0.973x ➖, 1↑ 0↓)
duckdb / vortex-compact (0.994x ➖, 0↑ 0↓)
duckdb / parquet (0.996x ➖, 0↑ 0↓)
Full attributed analysis
|
Benchmarks: Clickbench on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.915x ➖, 15↑ 0↓)
datafusion / parquet (0.912x ➖, 18↑ 1↓)
duckdb / vortex-file-compressed (0.990x ➖, 6↑ 2↓)
duckdb / parquet (0.993x ➖, 0↑ 0↓)
duckdb / duckdb (0.945x ➖, 7↑ 0↓)
Full attributed analysis
|
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Benchmarks: TPC-H SF=10 on S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (1.133x ➖, 1↑ 7↓)
datafusion / vortex-compact (1.058x ➖, 0↑ 5↓)
datafusion / parquet (0.977x ➖, 4↑ 5↓)
duckdb / vortex-file-compressed (1.022x ➖, 0↑ 3↓)
duckdb / vortex-compact (1.008x ➖, 0↑ 1↓)
duckdb / parquet (1.087x ➖, 0↑ 3↓)
Full attributed analysis
|
Benchmarks: CompressionVortex (geomean): 1.014x ➖ unknown / unknown (1.007x ➖, 0↑ 3↓)
|
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
Removes the temp
with_slotswith aslot_mutmethod this allows for arrays mutate the children inplace.