Skip to content

Commit 94de355

Browse files
prrao87prrao87
authored
docs: sync benchmark tables for lance-csr 0.3.1 (#69)
* docs: sync benchmark tables for lance-csr 0.3.1 * docs: sync benchmark tables from the mains-power runs * docs: pin benchmark tables to the merged runs
1 parent dbb8e9f commit 94de355

4 files changed

Lines changed: 128 additions & 113 deletions

File tree

‎README.md‎

Lines changed: 52 additions & 52 deletions
Original file line numberDiff line numberDiff line change
@@ -207,37 +207,37 @@ Over the 14 queries, lance-csr's speed relative to each engine:
207207

208208
| lance-csr mode | vs Neo4j | vs Kuzu | vs Ladybug |
209209
| --- | ---: | ---: | ---: |
210-
| In memory | 2.8× faster | 4.3× faster | 2.2× faster |
211-
| Lazy (Lance on disk) | 1.3× faster | 2.0× faster | 0.92× (slower) |
210+
| In memory | 3.9× faster | 6.1× faster | 3.2× faster |
211+
| Lazy (Lance on disk) | 1.8× faster | 2.8× faster | 1.3× faster |
212212

213213
<!-- /bench:ic:summary -->
214214

215215
Mean query latency in milliseconds (lower is better, fastest per query in bold):
216216

217217
<!-- bench:ic:table -->
218218

219-
| Query | lance-csr 0.2.0, in memory | lance-csr 0.2.0, lazy | Neo4j 2026.09.0 | Kuzu 0.11.3 | Ladybug 0.21.2 |
219+
| Query | lance-csr 0.3.1, in memory | lance-csr 0.3.1, lazy | Neo4j 2026.09.0 | Kuzu 0.11.3 | Ladybug 0.21.2 |
220220
| --- | ---: | ---: | ---: | ---: | ---: |
221-
| IC1 | 26.84 | 32.72 | 51.39 | **13.51** | n/a |
222-
| IC2 | **2.72** | 29.38 | 44.73 | 24.97 | 15.56 |
223-
| IC3 | **75.28** | 515.33 | 894.91 | 105.21 | 100.16 |
224-
| IC4 | **6.42** | 14.66 | 13.05 | 29.66 | 11.83 |
225-
| IC5 | **95.15** | 114.29 | 1,022.62 | 747.75 | 309.02 |
226-
| IC6 | 23.76 | **23.40** | 231.25 | 1,114.76 | 234.21 |
227-
| IC7 | 21.86 | 37.82 | **7.14** | 453.74 | 370.55 |
228-
| IC8 | **2.16** | 3.96 | 3.00 | 73.44 | 4.95 |
229-
| IC9 | **43.62** | 1,144.25 | 802.28 | 130.96 | 62.14 |
230-
| IC10 | 105.03 | 143.71 | 91.68 | 74.90 | **13.69** |
231-
| IC11 | 5.02 | 7.80 | 11.02 | 4.71 | **2.51** |
232-
| IC12 | 151.98 | 167.25 | 133.87 | **92.26** | 133.84 |
233-
| IC13 | 4.86 | 4.99 | **0.92** | 14.16 | 47.55 |
234-
| IC14 | **64.43** | 65.08 | 522.56 | 1,871.19 | n/a |
221+
| IC1 | 26.23 | 32.17 | 51.39 | **13.51** | n/a |
222+
| IC2 | **2.74** | 26.18 | 44.73 | 24.97 | 15.56 |
223+
| IC3 | **59.31** | 424.82 | 894.91 | 105.21 | 100.16 |
224+
| IC4 | **5.94** | 13.44 | 13.05 | 29.66 | 11.83 |
225+
| IC5 | **81.50** | 106.39 | 1,022.62 | 747.75 | 309.02 |
226+
| IC6 | **24.68** | 27.14 | 231.25 | 1,114.76 | 234.21 |
227+
| IC7 | 20.74 | 32.28 | **7.14** | 453.74 | 370.55 |
228+
| IC8 | **2.04** | 3.06 | 3.00 | 73.44 | 4.95 |
229+
| IC9 | **41.78** | 825.02 | 802.28 | 130.96 | 62.14 |
230+
| IC10 | 100.21 | 136.60 | 91.68 | 74.90 | **13.69** |
231+
| IC11 | 3.73 | 6.05 | 11.02 | 4.71 | **2.51** |
232+
| IC12 | **19.77** | 26.75 | 133.87 | 92.26 | 133.84 |
233+
| IC13 | 1.11 | 1.45 | **0.92** | 14.16 | 47.55 |
234+
| IC14 | **46.80** | 47.88 | 522.56 | 1,871.19 | n/a |
235235

236236
<!-- /bench:ic:table -->
237237

238238
<!-- bench:ic:notes -->
239239

240-
- Measured on 2026-10-24 on Apple M5, 24 GB, macOS 26.6.2, from the benchmark repository at commit [`c185123`](https://github.com/hyperdimensionalcomputingai/graph-benchmark-ldbc-complex/tree/c185123edd58b37b417bc4deb2fec643fdbd5e61). Every engine ran one after another with the repository's `run_benchmarks.sh` settings; see its README to reproduce the runs.
240+
- Run on Apple M5, 24 GB, macOS 26.6.2, from the benchmark repository at commit [`f3e63fd`](https://github.com/hyperdimensionalcomputingai/graph-benchmark-ldbc-complex/tree/f3e63fd2b7239a1364a864458f3e7ae07f0d833f). lance-csr was measured on 2026-10-27 and the other engines on 2026-10-24, on the same machine, with the repository's `run_benchmarks.sh` settings; see its README to reproduce the runs.
241241
- Ratios are geometric means of the per-query ratios, over the queries both engines answered correctly.
242242
- Ladybug 0.21.2: IC1 returned a wrong answer; IC14 failed (RuntimeError), so they are not timed.
243243

@@ -258,53 +258,53 @@ Over the 30 queries, lance-csr's speed relative to each engine:
258258

259259
| lance-csr mode | vs Neo4j | vs Kuzu | vs Ladybug | vs lance-graph |
260260
| --- | ---: | ---: | ---: | ---: |
261-
| In memory | 3.3× faster | 3.7× faster | 2.4× faster | 4.2× faster |
262-
| Lazy (Lance on disk) | 1.9× faster | 2.2× faster | 1.4× faster | 2.5× faster |
261+
| In memory | 3.6× faster | 4.1× faster | 2.6× faster | 4.6× faster |
262+
| Lazy (Lance on disk) | 2.1× faster | 2.3× faster | 1.5× faster | 2.6× faster |
263263

264264
<!-- /bench:q30:summary -->
265265

266266
Mean query latency in milliseconds (lower is better, fastest per query in bold):
267267

268268
<!-- bench:q30:table -->
269269

270-
| Query | lance-csr 0.2.0, in memory | lance-csr 0.2.0, lazy | Neo4j 2026.09.0 | Kuzu 0.11.3 | Ladybug 0.21.2 | lance-graph 0.5.4 |
270+
| Query | lance-csr 0.3.1, in memory | lance-csr 0.3.1, lazy | Neo4j 2026.09.0 | Kuzu 0.11.3 | Ladybug 0.21.2 | lance-graph 0.5.4 |
271271
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
272-
| Q1 | **0.74** | 1.18 | 1.28 | 1.73 | 1.22 | 1.30 |
273-
| Q2 | **0.58** | 1.41 | 3.26 | 1.32 | 1.12 | 2.34 |
274-
| Q3 | 0.69 | 0.83 | **0.66** | 1.03 | 0.79 | 1.86 |
275-
| Q4 | 0.54 | 0.83 | 2.66 | 0.85 | **0.39** | 2.98 |
276-
| Q5 | **0.82** | 2.24 | 2.78 | 3.44 | 4.00 | 2.04 |
277-
| Q6 | **0.47** | 0.75 | 2.60 | 0.66 | 0.79 | 0.73 |
278-
| Q7 | 1.09 | 1.13 | **0.43** | 27.73 | 25.11 | 15.68 |
279-
| Q8 | **0.68** | 1.82 | 7.29 | 2.72 | 3.38 | 1.14 |
280-
| Q9 | 0.86 | 1.38 | **0.76** | 1.73 | 1.34 | 1.89 |
281-
| Q10 | **0.77** | 0.98 | 3.79 | 1.55 | 1.78 | 26.69 |
282-
| Q11 | **1.59** | 1.99 | 10.28 | 7.96 | 4.63 | 3.08 |
283-
| Q12 | **0.84** | 7.77 | 2.55 | 17.16 | 7.84 | 18.68 |
284-
| Q13 | **0.80** | 1.06 | 6.65 | 42.66 | 5.84 | 9.62 |
285-
| Q14 | 0.55 | 0.69 | **0.46** | 1.56 | 0.90 | 2.91 |
286-
| Q15 | **0.56** | 0.78 | 2.37 | 2.38 | 1.16 | 3.02 |
287-
| Q16 | **0.70** | 1.16 | 1.26 | 1.78 | 1.04 | 4.33 |
288-
| Q17 | **0.98** | 1.44 | 3.09 | 2.62 | 1.74 | 2.30 |
289-
| Q18 | **0.70** | 1.17 | 3.11 | 1.51 | 1.22 | 1.76 |
290-
| Q19 | **0.98** | 1.61 | 4.86 | 11.66 | 3.43 | 19.93 |
291-
| Q20 | **2.12** | 19.59 | 421.92 | 13.25 | 14.15 | 3.13 |
292-
| Q21 | 0.77 | 0.94 | 0.51 | **0.44** | 0.50 | 1.46 |
293-
| Q22 | **0.71** | 0.93 | 2.66 | 23.34 | 7.46 | 13.41 |
294-
| Q23 | **0.74** | 1.22 | 3.13 | 1.19 | 0.79 | 2.42 |
295-
| Q24 | **0.69** | 1.07 | 0.72 | 1.23 | 0.81 | 1.75 |
296-
| Q25 | **0.80** | 1.24 | 2.86 | 1.45 | 1.27 | 1.47 |
297-
| Q26 | 0.75 | 1.10 | **0.65** | 3.37 | 1.15 | 2.93 |
298-
| Q27 | **0.87** | 1.33 | 2.22 | 17.96 | 3.71 | 24.84 |
299-
| Q28 | **1.39** | 2.31 | 2.37 | 1.51 | 1.78 | 2.50 |
300-
| Q29 | **0.69** | 1.03 | 2.62 | 1.02 | 0.72 | 2.96 |
301-
| Q30 | 35.36 | 39.70 | 1,359.98 | 146.52 | 97.77 | **34.31** |
272+
| Q1 | **0.66** | 1.13 | 1.28 | 1.73 | 1.22 | 1.30 |
273+
| Q2 | **0.51** | 1.33 | 3.26 | 1.32 | 1.12 | 2.34 |
274+
| Q3 | **0.62** | 0.77 | 0.66 | 1.03 | 0.79 | 1.86 |
275+
| Q4 | 0.48 | 0.75 | 2.66 | 0.85 | **0.39** | 2.98 |
276+
| Q5 | **0.77** | 2.23 | 2.78 | 3.44 | 4.00 | 2.04 |
277+
| Q6 | **0.41** | 0.71 | 2.60 | 0.66 | 0.79 | 0.73 |
278+
| Q7 | 0.98 | 1.05 | **0.43** | 27.73 | 25.11 | 15.68 |
279+
| Q8 | **0.61** | 1.75 | 7.29 | 2.72 | 3.38 | 1.14 |
280+
| Q9 | 0.81 | 1.32 | **0.76** | 1.73 | 1.34 | 1.89 |
281+
| Q10 | **0.70** | 0.91 | 3.79 | 1.55 | 1.78 | 26.69 |
282+
| Q11 | **1.47** | 1.89 | 10.28 | 7.96 | 4.63 | 3.08 |
283+
| Q12 | **0.69** | 6.80 | 2.55 | 17.16 | 7.84 | 18.68 |
284+
| Q13 | **0.74** | 0.99 | 6.65 | 42.66 | 5.84 | 9.62 |
285+
| Q14 | 0.48 | 0.65 | **0.46** | 1.56 | 0.90 | 2.91 |
286+
| Q15 | **0.51** | 0.70 | 2.37 | 2.38 | 1.16 | 3.02 |
287+
| Q16 | **0.62** | 1.06 | 1.26 | 1.78 | 1.04 | 4.33 |
288+
| Q17 | **0.87** | 1.37 | 3.09 | 2.62 | 1.74 | 2.30 |
289+
| Q18 | **0.63** | 1.07 | 3.11 | 1.51 | 1.22 | 1.76 |
290+
| Q19 | **0.90** | 1.33 | 4.86 | 11.66 | 3.43 | 19.93 |
291+
| Q20 | **2.05** | 16.45 | 421.92 | 13.25 | 14.15 | 3.13 |
292+
| Q21 | 0.70 | 0.79 | 0.51 | **0.44** | 0.50 | 1.46 |
293+
| Q22 | **0.65** | 0.88 | 2.66 | 23.34 | 7.46 | 13.41 |
294+
| Q23 | **0.66** | 1.18 | 3.13 | 1.19 | 0.79 | 2.42 |
295+
| Q24 | **0.63** | 1.00 | 0.72 | 1.23 | 0.81 | 1.75 |
296+
| Q25 | **0.72** | 1.13 | 2.86 | 1.45 | 1.27 | 1.47 |
297+
| Q26 | 0.67 | 1.02 | **0.65** | 3.37 | 1.15 | 2.93 |
298+
| Q27 | **0.80** | 1.24 | 2.22 | 17.96 | 3.71 | 24.84 |
299+
| Q28 | **1.36** | 2.25 | 2.37 | 1.51 | 1.78 | 2.50 |
300+
| Q29 | **0.62** | 0.97 | 2.62 | 1.02 | 0.72 | 2.96 |
301+
| Q30 | 43.67 | 47.50 | 1,359.98 | 146.52 | 97.77 | **34.31** |
302302

303303
<!-- /bench:q30:table -->
304304

305305
<!-- bench:q30:notes -->
306306

307-
- Measured on 2026-10-24 on Apple M5, 24 GB, macOS 26.6.2, from the benchmark repository at commit [`6260689`](https://github.com/hyperdimensionalcomputingai/graph-benchmark-ldbc/tree/6260689aa1fba4eeb4e99958430386b2fcd13f71). Every engine ran one after another with the repository's `run_benchmarks.sh` settings; see its README to reproduce the runs.
307+
- Run on Apple M5, 24 GB, macOS 26.6.2, from the benchmark repository at commit [`b649bc1`](https://github.com/hyperdimensionalcomputingai/graph-benchmark-ldbc/tree/b649bc166cb5a04a542882fc070d9c4d83507c65). lance-csr was measured on 2026-10-27 and the other engines on 2026-10-24, on the same machine, with the repository's `run_benchmarks.sh` settings; see its README to reproduce the runs.
308308
- Ratios are geometric means of the per-query ratios, over the queries both engines answered correctly.
309309

310310
<!-- /bench:q30:notes -->

‎_typos.toml‎

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,3 @@
1+
[default]
2+
# Git commit hashes, such as the benchmark commits pinned in docs/sync_benchmarks.py.
3+
extend-ignore-re = ["\\b[0-9a-f]{7,40}\\b"]

‎docs/benchmarks.md‎

Lines changed: 52 additions & 52 deletions
Original file line numberDiff line numberDiff line change
@@ -42,37 +42,37 @@ Over the 14 queries, lance-csr's speed relative to each engine:
4242

4343
| lance-csr mode | vs Neo4j | vs Kuzu | vs Ladybug |
4444
| --- | ---: | ---: | ---: |
45-
| In memory | 2.8× faster | 4.3× faster | 2.2× faster |
46-
| Lazy (Lance on disk) | 1.3× faster | 2.0× faster | 0.92× (slower) |
45+
| In memory | 3.9× faster | 6.1× faster | 3.2× faster |
46+
| Lazy (Lance on disk) | 1.8× faster | 2.8× faster | 1.3× faster |
4747

4848
<!-- /bench:ic:summary -->
4949

5050
??? info "Per-query mean latency (ms, lower is better, fastest in bold)"
5151

5252
<!-- bench:ic:table -->
5353

54-
| Query | lance-csr 0.2.0, in memory | lance-csr 0.2.0, lazy | Neo4j 2026.09.0 | Kuzu 0.11.3 | Ladybug 0.21.2 |
54+
| Query | lance-csr 0.3.1, in memory | lance-csr 0.3.1, lazy | Neo4j 2026.09.0 | Kuzu 0.11.3 | Ladybug 0.21.2 |
5555
| --- | ---: | ---: | ---: | ---: | ---: |
56-
| IC1 | 26.84 | 32.72 | 51.39 | **13.51** | n/a |
57-
| IC2 | **2.72** | 29.38 | 44.73 | 24.97 | 15.56 |
58-
| IC3 | **75.28** | 515.33 | 894.91 | 105.21 | 100.16 |
59-
| IC4 | **6.42** | 14.66 | 13.05 | 29.66 | 11.83 |
60-
| IC5 | **95.15** | 114.29 | 1,022.62 | 747.75 | 309.02 |
61-
| IC6 | 23.76 | **23.40** | 231.25 | 1,114.76 | 234.21 |
62-
| IC7 | 21.86 | 37.82 | **7.14** | 453.74 | 370.55 |
63-
| IC8 | **2.16** | 3.96 | 3.00 | 73.44 | 4.95 |
64-
| IC9 | **43.62** | 1,144.25 | 802.28 | 130.96 | 62.14 |
65-
| IC10 | 105.03 | 143.71 | 91.68 | 74.90 | **13.69** |
66-
| IC11 | 5.02 | 7.80 | 11.02 | 4.71 | **2.51** |
67-
| IC12 | 151.98 | 167.25 | 133.87 | **92.26** | 133.84 |
68-
| IC13 | 4.86 | 4.99 | **0.92** | 14.16 | 47.55 |
69-
| IC14 | **64.43** | 65.08 | 522.56 | 1,871.19 | n/a |
56+
| IC1 | 26.23 | 32.17 | 51.39 | **13.51** | n/a |
57+
| IC2 | **2.74** | 26.18 | 44.73 | 24.97 | 15.56 |
58+
| IC3 | **59.31** | 424.82 | 894.91 | 105.21 | 100.16 |
59+
| IC4 | **5.94** | 13.44 | 13.05 | 29.66 | 11.83 |
60+
| IC5 | **81.50** | 106.39 | 1,022.62 | 747.75 | 309.02 |
61+
| IC6 | **24.68** | 27.14 | 231.25 | 1,114.76 | 234.21 |
62+
| IC7 | 20.74 | 32.28 | **7.14** | 453.74 | 370.55 |
63+
| IC8 | **2.04** | 3.06 | 3.00 | 73.44 | 4.95 |
64+
| IC9 | **41.78** | 825.02 | 802.28 | 130.96 | 62.14 |
65+
| IC10 | 100.21 | 136.60 | 91.68 | 74.90 | **13.69** |
66+
| IC11 | 3.73 | 6.05 | 11.02 | 4.71 | **2.51** |
67+
| IC12 | **19.77** | 26.75 | 133.87 | 92.26 | 133.84 |
68+
| IC13 | 1.11 | 1.45 | **0.92** | 14.16 | 47.55 |
69+
| IC14 | **46.80** | 47.88 | 522.56 | 1,871.19 | n/a |
7070

7171
<!-- /bench:ic:table -->
7272

7373
<!-- bench:ic:notes -->
7474

75-
- Measured on 2026-10-24 on Apple M5, 24 GB, macOS 26.6.2, from the benchmark repository at commit [`c185123`](https://github.com/hyperdimensionalcomputingai/graph-benchmark-ldbc-complex/tree/c185123edd58b37b417bc4deb2fec643fdbd5e61). Every engine ran one after another with the repository's `run_benchmarks.sh` settings; see its README to reproduce the runs.
75+
- Run on Apple M5, 24 GB, macOS 26.6.2, from the benchmark repository at commit [`f3e63fd`](https://github.com/hyperdimensionalcomputingai/graph-benchmark-ldbc-complex/tree/f3e63fd2b7239a1364a864458f3e7ae07f0d833f). lance-csr was measured on 2026-10-27 and the other engines on 2026-10-24, on the same machine, with the repository's `run_benchmarks.sh` settings; see its README to reproduce the runs.
7676
- Ratios are geometric means of the per-query ratios, over the queries both engines answered correctly.
7777
- Ladybug 0.21.2: IC1 returned a wrong answer; IC14 failed (RuntimeError), so they are not timed.
7878

@@ -99,53 +99,53 @@ Over the 30 queries, lance-csr's speed relative to each engine:
9999

100100
| lance-csr mode | vs Neo4j | vs Kuzu | vs Ladybug | vs lance-graph |
101101
| --- | ---: | ---: | ---: | ---: |
102-
| In memory | 3.3× faster | 3.7× faster | 2.4× faster | 4.2× faster |
103-
| Lazy (Lance on disk) | 1.9× faster | 2.2× faster | 1.4× faster | 2.5× faster |
102+
| In memory | 3.6× faster | 4.1× faster | 2.6× faster | 4.6× faster |
103+
| Lazy (Lance on disk) | 2.1× faster | 2.3× faster | 1.5× faster | 2.6× faster |
104104

105105
<!-- /bench:q30:summary -->
106106

107107
??? info "Per-query mean latency (ms, lower is better, fastest in bold)"
108108

109109
<!-- bench:q30:table -->
110110

111-
| Query | lance-csr 0.2.0, in memory | lance-csr 0.2.0, lazy | Neo4j 2026.09.0 | Kuzu 0.11.3 | Ladybug 0.21.2 | lance-graph 0.5.4 |
111+
| Query | lance-csr 0.3.1, in memory | lance-csr 0.3.1, lazy | Neo4j 2026.09.0 | Kuzu 0.11.3 | Ladybug 0.21.2 | lance-graph 0.5.4 |
112112
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
113-
| Q1 | **0.74** | 1.18 | 1.28 | 1.73 | 1.22 | 1.30 |
114-
| Q2 | **0.58** | 1.41 | 3.26 | 1.32 | 1.12 | 2.34 |
115-
| Q3 | 0.69 | 0.83 | **0.66** | 1.03 | 0.79 | 1.86 |
116-
| Q4 | 0.54 | 0.83 | 2.66 | 0.85 | **0.39** | 2.98 |
117-
| Q5 | **0.82** | 2.24 | 2.78 | 3.44 | 4.00 | 2.04 |
118-
| Q6 | **0.47** | 0.75 | 2.60 | 0.66 | 0.79 | 0.73 |
119-
| Q7 | 1.09 | 1.13 | **0.43** | 27.73 | 25.11 | 15.68 |
120-
| Q8 | **0.68** | 1.82 | 7.29 | 2.72 | 3.38 | 1.14 |
121-
| Q9 | 0.86 | 1.38 | **0.76** | 1.73 | 1.34 | 1.89 |
122-
| Q10 | **0.77** | 0.98 | 3.79 | 1.55 | 1.78 | 26.69 |
123-
| Q11 | **1.59** | 1.99 | 10.28 | 7.96 | 4.63 | 3.08 |
124-
| Q12 | **0.84** | 7.77 | 2.55 | 17.16 | 7.84 | 18.68 |
125-
| Q13 | **0.80** | 1.06 | 6.65 | 42.66 | 5.84 | 9.62 |
126-
| Q14 | 0.55 | 0.69 | **0.46** | 1.56 | 0.90 | 2.91 |
127-
| Q15 | **0.56** | 0.78 | 2.37 | 2.38 | 1.16 | 3.02 |
128-
| Q16 | **0.70** | 1.16 | 1.26 | 1.78 | 1.04 | 4.33 |
129-
| Q17 | **0.98** | 1.44 | 3.09 | 2.62 | 1.74 | 2.30 |
130-
| Q18 | **0.70** | 1.17 | 3.11 | 1.51 | 1.22 | 1.76 |
131-
| Q19 | **0.98** | 1.61 | 4.86 | 11.66 | 3.43 | 19.93 |
132-
| Q20 | **2.12** | 19.59 | 421.92 | 13.25 | 14.15 | 3.13 |
133-
| Q21 | 0.77 | 0.94 | 0.51 | **0.44** | 0.50 | 1.46 |
134-
| Q22 | **0.71** | 0.93 | 2.66 | 23.34 | 7.46 | 13.41 |
135-
| Q23 | **0.74** | 1.22 | 3.13 | 1.19 | 0.79 | 2.42 |
136-
| Q24 | **0.69** | 1.07 | 0.72 | 1.23 | 0.81 | 1.75 |
137-
| Q25 | **0.80** | 1.24 | 2.86 | 1.45 | 1.27 | 1.47 |
138-
| Q26 | 0.75 | 1.10 | **0.65** | 3.37 | 1.15 | 2.93 |
139-
| Q27 | **0.87** | 1.33 | 2.22 | 17.96 | 3.71 | 24.84 |
140-
| Q28 | **1.39** | 2.31 | 2.37 | 1.51 | 1.78 | 2.50 |
141-
| Q29 | **0.69** | 1.03 | 2.62 | 1.02 | 0.72 | 2.96 |
142-
| Q30 | 35.36 | 39.70 | 1,359.98 | 146.52 | 97.77 | **34.31** |
113+
| Q1 | **0.66** | 1.13 | 1.28 | 1.73 | 1.22 | 1.30 |
114+
| Q2 | **0.51** | 1.33 | 3.26 | 1.32 | 1.12 | 2.34 |
115+
| Q3 | **0.62** | 0.77 | 0.66 | 1.03 | 0.79 | 1.86 |
116+
| Q4 | 0.48 | 0.75 | 2.66 | 0.85 | **0.39** | 2.98 |
117+
| Q5 | **0.77** | 2.23 | 2.78 | 3.44 | 4.00 | 2.04 |
118+
| Q6 | **0.41** | 0.71 | 2.60 | 0.66 | 0.79 | 0.73 |
119+
| Q7 | 0.98 | 1.05 | **0.43** | 27.73 | 25.11 | 15.68 |
120+
| Q8 | **0.61** | 1.75 | 7.29 | 2.72 | 3.38 | 1.14 |
121+
| Q9 | 0.81 | 1.32 | **0.76** | 1.73 | 1.34 | 1.89 |
122+
| Q10 | **0.70** | 0.91 | 3.79 | 1.55 | 1.78 | 26.69 |
123+
| Q11 | **1.47** | 1.89 | 10.28 | 7.96 | 4.63 | 3.08 |
124+
| Q12 | **0.69** | 6.80 | 2.55 | 17.16 | 7.84 | 18.68 |
125+
| Q13 | **0.74** | 0.99 | 6.65 | 42.66 | 5.84 | 9.62 |
126+
| Q14 | 0.48 | 0.65 | **0.46** | 1.56 | 0.90 | 2.91 |
127+
| Q15 | **0.51** | 0.70 | 2.37 | 2.38 | 1.16 | 3.02 |
128+
| Q16 | **0.62** | 1.06 | 1.26 | 1.78 | 1.04 | 4.33 |
129+
| Q17 | **0.87** | 1.37 | 3.09 | 2.62 | 1.74 | 2.30 |
130+
| Q18 | **0.63** | 1.07 | 3.11 | 1.51 | 1.22 | 1.76 |
131+
| Q19 | **0.90** | 1.33 | 4.86 | 11.66 | 3.43 | 19.93 |
132+
| Q20 | **2.05** | 16.45 | 421.92 | 13.25 | 14.15 | 3.13 |
133+
| Q21 | 0.70 | 0.79 | 0.51 | **0.44** | 0.50 | 1.46 |
134+
| Q22 | **0.65** | 0.88 | 2.66 | 23.34 | 7.46 | 13.41 |
135+
| Q23 | **0.66** | 1.18 | 3.13 | 1.19 | 0.79 | 2.42 |
136+
| Q24 | **0.63** | 1.00 | 0.72 | 1.23 | 0.81 | 1.75 |
137+
| Q25 | **0.72** | 1.13 | 2.86 | 1.45 | 1.27 | 1.47 |
138+
| Q26 | 0.67 | 1.02 | **0.65** | 3.37 | 1.15 | 2.93 |
139+
| Q27 | **0.80** | 1.24 | 2.22 | 17.96 | 3.71 | 24.84 |
140+
| Q28 | **1.36** | 2.25 | 2.37 | 1.51 | 1.78 | 2.50 |
141+
| Q29 | **0.62** | 0.97 | 2.62 | 1.02 | 0.72 | 2.96 |
142+
| Q30 | 43.67 | 47.50 | 1,359.98 | 146.52 | 97.77 | **34.31** |
143143

144144
<!-- /bench:q30:table -->
145145

146146
<!-- bench:q30:notes -->
147147

148-
- Measured on 2026-10-24 on Apple M5, 24 GB, macOS 26.6.2, from the benchmark repository at commit [`6260689`](https://github.com/hyperdimensionalcomputingai/graph-benchmark-ldbc/tree/6260689aa1fba4eeb4e99958430386b2fcd13f71). Every engine ran one after another with the repository's `run_benchmarks.sh` settings; see its README to reproduce the runs.
148+
- Run on Apple M5, 24 GB, macOS 26.6.2, from the benchmark repository at commit [`b649bc1`](https://github.com/hyperdimensionalcomputingai/graph-benchmark-ldbc/tree/b649bc166cb5a04a542882fc070d9c4d83507c65). lance-csr was measured on 2026-10-27 and the other engines on 2026-10-24, on the same machine, with the repository's `run_benchmarks.sh` settings; see its README to reproduce the runs.
149149
- Ratios are geometric means of the per-query ratios, over the queries both engines answered correctly.
150150

151151
<!-- /bench:q30:notes -->

0 commit comments

Comments
 (0)