perf: optimize varbinview construction
#8146Comparing
claude/avx512-decode-benchmark-wRhkZ
(fc6b839
) withdevelop
(73454db
)+11%
- Improvement
- 7
- Untouched
- 1268
Unknown Walltime execution environment detected
Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data. For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.Different runtime environments detected
Some benchmarks with significant performance changes were compared across different runtime environments, which may affect the accuracy of the results. Learn more about regression causes.Benchmarks
1275 total
decompress_fsst[(10000, 16, 4)]
encodings/fsst/benches/fsst_compress.rs+13%
579.4 µs514.8 µs
chunked_canonicalize_into[(10, 10000, 16, 4)]
encodings/fsst/benches/fsst_compress.rs+12%
5.9 ms5.3 ms
chunked_into_canonical[(10, 10000, 16, 4)]
encodings/fsst/benches/fsst_compress.rs+12%
5.9 ms5.3 ms
fsst_decompress_string
vortex/benches/single_encoding_throughput.rs+12%
3.5 ms3.1 ms
decompress_fsst[(10000, 16, 8)]
encodings/fsst/benches/fsst_compress.rs+11%
631.8 µs567.2 µs
decompress_fsst[(10000, 4, 4)]
encodings/fsst/benches/fsst_compress.rs+10%
388.3 µs352.7 µs
canonicalize_compare[(10000, 16, 4)]
encodings/fsst/benches/fsst_compress.rs+10%
704.7 µs640.5 µs
decompress_fsst[(10000, 4, 8)]
encodings/fsst/benches/fsst_compress.rs+10%
398.1 µs362 µs
chunked_into_canonical[(10, 10000, 4, 4)]
encodings/fsst/benches/fsst_compress.rs+10%
4 ms3.6 ms
chunked_canonicalize_into[(10, 10000, 4, 4)]
encodings/fsst/benches/fsst_compress.rs+10%
4 ms3.6 ms
decompress_fsst[(1000, 16, 4)]
encodings/fsst/benches/fsst_compress.rs+10%
79.8 µs72.8 µs
canonicalize_compare[(10000, 16, 8)]
encodings/fsst/benches/fsst_compress.rs+9%
757.5 µs692.8 µs
decompress_fsst[(1000, 16, 8)]
encodings/fsst/benches/fsst_compress.rs+9%
85.1 µs78 µs
chunked_into_canonical[(1000, 100, 16, 4)]
encodings/fsst/benches/fsst_compress.rs+8%
9.6 ms8.9 ms
canonicalize_compare[(10000, 4, 8)]
encodings/fsst/benches/fsst_compress.rs+7%
525.4 µs488.9 µs
canonicalize_compare[(10000, 4, 4)]
encodings/fsst/benches/fsst_compress.rs+7%
515.5 µs479.7 µs
decompress_fsst[(1000, 4, 8)]
encodings/fsst/benches/fsst_compress.rs+7%
62.1 µs58.1 µs
decompress_fsst[(1000, 4, 4)]
encodings/fsst/benches/fsst_compress.rs+7%
60.7 µs56.8 µs
chunked_canonicalize_into[(1000, 100, 16, 4)]
encodings/fsst/benches/fsst_compress.rs+7%
9.1 ms8.5 ms
chunked_into_canonical[(1000, 100, 16, 16)]
encodings/fsst/benches/fsst_compress.rs+7%
11.4 ms10.7 ms
chunked_canonicalize_into[(1000, 100, 16, 16)]
encodings/fsst/benches/fsst_compress.rs+6%
10.9 ms10.3 ms
chunked_into_canonical[(1000, 100, 16, 64)]
encodings/fsst/benches/fsst_compress.rs+6%
13.9 ms13.1 ms
eq_canonicalize_low_match
encodings/fsst/benches/fsst_url_compare.rs+6%
13.4 ms12.6 ms
eq_canonicalize_high_match
encodings/fsst/benches/fsst_url_compare.rs+6%
13.5 ms12.7 ms
chunked_canonicalize_into[(1000, 100, 16, 64)]
encodings/fsst/benches/fsst_compress.rs+6%
13.3 ms12.6 ms
Commits
Click on a commit to change the comparison rangeBase
develop
73454db
+0.23%
Speed up VarBinView build_views single-buffer canonicalization
50b0e93
1 day ago
by joseph-isaacs+11.94%
Optimize build_views single-buffer hot loop write path
76e17aa
15 hours ago
by claude-0.93%
test: cover build_views single-buffer fast path
84159af
11 hours ago
by claude-1.53%
test: verify build_views fast path large u32 offsets
a95257e
11 hours ago
by claude0%
test: slow gated FSST canonicalize offsets-overflow regression
10f6f01
10 hours ago
by claude+1.71%
test: replace FSST overflow test with direct build_views rollover test
fc6b839
10 hours ago
by claude