Skip to content

Commit 398f9f8

Browse files
docs: update query benchmarks (3.4.1) (#646)
* docs: update query benchmarks (3.4.1) * fix(docs): add explanatory note for 3.4.1 diffImpact +45% delta (#646) * fix(docs): restore 3.4.1 raw totals lost in merge, add engine parity note (#647) The merge with main reverted raw totals, phase breakdowns, incremental rebuild rows, query rows, and the JSON blob back to 3.4.0 values. Restores the correct 3.4.1 data and adds a note referencing #649 for the native/WASM node count divergence (10,982 vs 10,956). * fix(docs): add explanatory note for 3.4.1 diffImpact +45% delta (#646) * fix(docs): restore 3.4.1 query benchmark data and update latest results (#646) The previous fix commit accidentally removed all 3.4.1 data instead of adding the explanatory note. This restores the summary table rows, "Latest results" section, and JSON block for 3.4.1. --------- Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: carlos-alm <127798846+carlos-alm@users.noreply.github.com>
1 parent aa6499a commit 398f9f8

1 file changed

Lines changed: 67 additions & 15 deletions

File tree

generated/benchmarks/QUERY-BENCHMARKS.md

Lines changed: 67 additions & 15 deletions
Original file line numberDiff line numberDiff line change
@@ -5,6 +5,8 @@ Latencies are median over 5 runs. Hub target = most-connected node.
55

66
| Version | Engine | fnDeps d1 | fnDeps d3 | fnDeps d5 | fnImpact d1 | fnImpact d3 | fnImpact d5 | diffImpact |
77
|---------|--------|----------:|----------:|----------:|------------:|------------:|------------:|-----------:|
8+
| 3.4.1 | native | 8.9 ↑5% | 8.9 ↑5% | 8.8 ↑5% | 3.2 ~ | 3.1 ↓3% | 3.1 ↓3% | 8.1ms ↑45% |
9+
| 3.4.1 | wasm | 9.1 ↑7% | 9.2 ↑7% | 9.1 ↑7% | 3.3 ~ | 3.2 ↓3% | 3.2 ↓3% | 7.1ms ↑45% |
810
| 3.4.0 | native | 8.5 ↑136% | 8.5 ↑136% | 8.4 ↑140% | 3.2 ↑28% | 3.2 ↑23% | 3.2 ↑28% | 5.6ms ↓20% |
911
| 3.4.0 | wasm | 8.5 ↑143% | 8.6 ↑139% | 8.5 ↑143% | 3.3 ↑38% | 3.3 ↑38% | 3.3 ↑38% | 4.9ms ↑2% |
1012
| 3.3.1 | native | 3.6 ↑157% | 3.6 ↑177% | 3.5 ↑169% | 2.5 ↑92% | 2.6 ↑100% | 2.5 ↑92% | 7ms ↓20% |
@@ -36,48 +38,98 @@ Latencies are median over 5 runs. Hub target = most-connected node.
3638

3739
### Latest results
3840

39-
**Version:** 3.4.0 | **Date:** 2026-03-26
41+
**Version:** 3.4.1 | **Date:** 2026-03-27
4042

4143
#### Native (Rust)
4244

43-
**Targets:** hub=`buildGraph`, mid=`rule`, leaf=`docs`
45+
**Targets:** hub=`buildGraph`, mid=`noTests`, leaf=`docs`
4446

4547
| Metric | Value |
4648
|--------|------:|
47-
| fnDeps depth 1 | 8.5ms |
48-
| fnDeps depth 3 | 8.5ms |
49-
| fnDeps depth 5 | 8.4ms |
49+
| fnDeps depth 1 | 8.9ms |
50+
| fnDeps depth 3 | 8.9ms |
51+
| fnDeps depth 5 | 8.8ms |
5052
| fnImpact depth 1 | 3.2ms |
51-
| fnImpact depth 3 | 3.2ms |
52-
| fnImpact depth 5 | 3.2ms |
53-
| diffImpact latency | 5.6ms |
53+
| fnImpact depth 3 | 3.1ms |
54+
| fnImpact depth 5 | 3.1ms |
55+
| diffImpact latency | 8.1ms |
5456
| diffImpact affected functions | 0 |
5557
| diffImpact affected files | 0 |
5658

5759
#### WASM
5860

59-
**Targets:** hub=`buildGraph`, mid=`rule`, leaf=`docs`
61+
**Targets:** hub=`buildGraph`, mid=`noTests`, leaf=`docs`
6062

6163
| Metric | Value |
6264
|--------|------:|
63-
| fnDeps depth 1 | 8.5ms |
64-
| fnDeps depth 3 | 8.6ms |
65-
| fnDeps depth 5 | 8.5ms |
65+
| fnDeps depth 1 | 9.1ms |
66+
| fnDeps depth 3 | 9.2ms |
67+
| fnDeps depth 5 | 9.1ms |
6668
| fnImpact depth 1 | 3.3ms |
67-
| fnImpact depth 3 | 3.3ms |
68-
| fnImpact depth 5 | 3.3ms |
69-
| diffImpact latency | 4.9ms |
69+
| fnImpact depth 3 | 3.2ms |
70+
| fnImpact depth 5 | 3.2ms |
71+
| diffImpact latency | 7.1ms |
7072
| diffImpact affected functions | 0 |
7173
| diffImpact affected files | 0 |
7274

7375
<!-- NOTES_START -->
76+
**Note (3.4.1):** The ↑45% diffImpact delta is inflated by 3.4.0 being an unusually low baseline (5.6ms/4.9ms). The new absolute values (8.1ms native, 7.1ms wasm) fall within the historical range (e.g. 3.1.3: 8.3ms, 3.3.0: 8.8ms). The mid-query target also changed from `rule` to `noTests`, which may affect diffImpact scope. No engine regression — fnDeps grew only 5-7% (consistent with codebase expansion) and fnImpact is flat-to-slightly-down.
77+
7478
**Note (3.4.0):** The ↑136-143% fnDeps deltas for 3.4.0 vs 3.3.1 reflect codebase growth — `buildGraph` has significantly more edges in this release (new extractors, refactored domain/features layers). The mid-query target also changed from `db` to `rule`. There is no engine regression — native `diffImpact` improved 20% in the same release.
7579

7680
**Note (3.3.1):** The ↑157-192% fnDeps/fnImpact deltas for 3.3.1 vs 3.3.0 are not comparable. PR #528 changed the hub target from auto-selected `src/types.ts` (shallow type-barrel) to pinned `buildGraph` (deep orchestration function with 2-3x more edges). There is no engine regression — `diffImpact` improved 20-44% in the same release. Future version comparisons (3.3.1+) are stable and meaningful.
7781
<!-- NOTES_END -->
7882

7983
<!-- QUERY_BENCHMARK_DATA
8084
[
85+
{
86+
"version": "3.4.1",
87+
"date": "2026-03-27",
88+
"wasm": {
89+
"targets": {
90+
"hub": "buildGraph",
91+
"mid": "noTests",
92+
"leaf": "docs"
93+
},
94+
"fnDeps": {
95+
"depth1Ms": 9.1,
96+
"depth3Ms": 9.2,
97+
"depth5Ms": 9.1
98+
},
99+
"fnImpact": {
100+
"depth1Ms": 3.3,
101+
"depth3Ms": 3.2,
102+
"depth5Ms": 3.2
103+
},
104+
"diffImpact": {
105+
"latencyMs": 7.1,
106+
"affectedFunctions": 0,
107+
"affectedFiles": 0
108+
}
109+
},
110+
"native": {
111+
"targets": {
112+
"hub": "buildGraph",
113+
"mid": "noTests",
114+
"leaf": "docs"
115+
},
116+
"fnDeps": {
117+
"depth1Ms": 8.9,
118+
"depth3Ms": 8.9,
119+
"depth5Ms": 8.8
120+
},
121+
"fnImpact": {
122+
"depth1Ms": 3.2,
123+
"depth3Ms": 3.1,
124+
"depth5Ms": 3.1
125+
},
126+
"diffImpact": {
127+
"latencyMs": 8.1,
128+
"affectedFunctions": 0,
129+
"affectedFiles": 0
130+
}
131+
}
132+
},
81133
{
82134
"version": "3.4.0",
83135
"date": "2026-03-26",

0 commit comments

Comments
 (0)