Avatar for the vortex-data user
vortex-data
vortex
BlogDocsChangelog

bench(fsst): add alloc-free decompress+match baselines for fair comparison

#6903
Comparing
claude/check-listing-L7l0k
(
a3c14d4
) with
develop
(
a1a9d77
)
CodSpeed Performance Gauge
×3.4
Improvement
2
Untouched
992
New
118
Skipped
1515

Benchmarks

1515 total
null_count_run_end[(100000, 256, 0.5)]
encodings/runend/benches/run_end_null_count.rs
Skipped
12.9 KB*
pco_canonical[(10000, 1.0)]
encodings/pco/benches/pco.rs
Skipped
128 B*
decompress[u64, (100000, 1024)]
encodings/runend/benches/run_end_compress.rs
Skipped
776.2 KB*
pco_canonical[(50000, 1.0)]
encodings/pco/benches/pco.rs
Skipped
128 B*
null_count_run_end[(10000, 1024, 0.5)]
encodings/runend/benches/run_end_null_count.rs
Skipped
1.8 KB*
null_count_run_end[(10000, 16, 0.1)]
encodings/runend/benches/run_end_null_count.rs
Skipped
1.9 KB*
take_indices[(10000, 256, 0.03)]
encodings/runend/benches/run_end_filter.rs
Skipped
5.7 KB*
decompress[u64, (1000000, 256)]
encodings/runend/benches/run_end_compress.rs
Skipped
7.6 MB*
pco_canonical[(100000, 0.5)]
encodings/pco/benches/pco.rs
Skipped
128 B*
take_indices[(1000, 4, 0.03)]
encodings/runend/benches/run_end_filter.rs
Skipped
1.4 KB*
decompress[u16, (1000, 256)]
encodings/runend/benches/run_end_compress.rs
Skipped
1.7 KB*
filter_runend[(10000, 4, 0.01)]
encodings/runend/benches/run_end_filter.rs
Skipped
23.1 KB*
take_indices[(10000, 1024)]
encodings/runend/benches/run_end_compress.rs
Skipped
1.9 KB*
take_indices[(1000, 16)]
encodings/runend/benches/run_end_compress.rs
Skipped
2.1 KB*
filter_runend[(10000, 256, 0.01)]
encodings/runend/benches/run_end_filter.rs
Skipped
1 KB*
filter_runend[(1000, 16, 0.03)]
encodings/runend/benches/run_end_filter.rs
Skipped
1.3 KB*
decompress[u64, (10000, 16)]
encodings/runend/benches/run_end_compress.rs
Skipped
78.3 KB*
take_indices[(10000, 256)]
encodings/runend/benches/run_end_compress.rs
Skipped
2 KB*
null_count_run_end[(100000, 256, 0.1)]
encodings/runend/benches/run_end_null_count.rs
Skipped
12.9 KB*
null_count_run_end[(10000, 256, 0.5)]
encodings/runend/benches/run_end_null_count.rs
Skipped
1.9 KB*
take_indices[(100000, 4)]
encodings/runend/benches/run_end_compress.rs
Skipped
99.5 KB*
decompress[u8, (10000, 4)]
encodings/runend/benches/run_end_compress.rs
Skipped
10 KB*
filter_runend[(1000, 4, 0.005)]
encodings/runend/benches/run_end_filter.rs
Skipped
2.9 KB*
filter_runend[(1000, 256, 0.005)]
encodings/runend/benches/run_end_filter.rs
Skipped
776 B*
null_count_run_end[(10000, 16, 0.5)]
encodings/runend/benches/run_end_null_count.rs
Skipped
1.9 KB*

Commits

Click on a commit to change the comparison range
Base
develop
a1a9d77
-0.05%
bench(fsst): add alloc-free decompress+match baselines for fair comparison
21c193a
1 day ago
by claude
-0.04%
bench(fsst): add hybrid DFAs and external crate benchmarks
f365e7c
1 day ago
by claude
+0.04%
feat(fsst): upgrade contains DFA to fused u8 table and add decompress-only benchmarks
90f90a5
1 day ago
by claude
0%
bench(fsst): add vortex array LIKE kernel and decompress-only benchmarks
38298c3
1 day ago
by claude
0%
perf(fsst): optimize LIKE kernel — shift prefix DFA, remove clone, drop early-exit
7fd52f1
1 day ago
by claude
0%
perf(fsst): replace collect_bool with inline u64 word packing in LIKE kernel
d68695b
1 day ago
by claude
0%
perf(fsst): batch offsets + iterator-based DFA with early-exit
c8a6418
1 day ago
by claude
+0.01%
perf(fsst): branchless DFA, running offset, iterator-based matching
1205017
1 day ago
by claude
×3.4
uxiwp
3f95753
17 hours ago
by joseph-isaacs
+0.05%
uxiwp
a3c14d4
16 hours ago
by joseph-isaacs
© 2026 CodSpeed Technology
Home Terms Privacy Docs