You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@doris.apache.org by "zhiqiang-hhhh (via GitHub)" <gi...@apache.org> on 2026/01/16 08:38:01 UTC
[PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
zhiqiang-hhhh opened a new pull request, #59968:
URL: https://github.com/apache/doris/pull/59968
### What problem does this PR solve?
Currently if input rows of ann topn search is less than 30% of the segment, we will fall back to brute force search for a better recall.
This pr intruduces a new session var ann_topn_min_input_rows_ratio to make threshold configurable.
Related PR: #xxx
Problem Summary:
### Release note
None
### Check List (For Author)
- Test <!-- At least one of them must be included. -->
- [ ] Regression test
- [ ] Unit Test
- [ ] Manual test (add detailed scripts or steps below)
- [ ] No need to test or manual test. Explain why:
- [ ] This is a refactor/code format and no logic has been changed.
- [ ] Previous test can cover this change.
- [ ] No code files have been changed.
- [ ] Other reason <!-- Add your reason? -->
- Behavior changed:
- [ ] No.
- [ ] Yes. <!-- Explain the behavior change -->
- Does this need documentation?
- [ ] No.
- [ ] Yes. <!-- Add document PR link here. eg: https://github.com/apache/doris-website/pull/1214 -->
### Check List (For Reviewer who merge this PR)
- [ ] Confirm the release note
- [ ] Confirm test cases
- [ ] Confirm document
- [ ] Add branch pick label <!-- Add branch pick label that this PR should merge into -->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "zhiqiang-hhhh (via GitHub)" <gi...@apache.org>.
zhiqiang-hhhh commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-3758778958
run buildall
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-3763516323
# Cloud UT Coverage Report
Increment line coverage ` ` :tada:
[Increment coverage report](http://coverage.selectdb-in.cc/coverage/43e1e44a64ab2448408c6da658809b9120ff5ffb_43e1e44a64ab2448408c6da658809b9120ff5ffb_cloud/increment_report/index.html)
[Complete coverage report](http://coverage.selectdb-in.cc/coverage/43e1e44a64ab2448408c6da658809b9120ff5ffb_43e1e44a64ab2448408c6da658809b9120ff5ffb_cloud/report/index.html)
| Category | Coverage |
|-------------------|--------------------|
| Function Coverage | 79.52% (1786/2246) |
| Line Coverage | 64.74% (31798/49114) |
| Region Coverage | 65.43% (15835/24201) |
| Branch Coverage | 55.98% (8413/15028) |
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] closed pull request #59968: [opt](ann index) Introduce ann_topn_min_input_rows_ratio
URL: https://github.com/apache/doris/pull/59968
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-3764845679
PR approved by at least one committer and no changes requested.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-3764845717
PR approved by anyone and no changes requested.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-3759487269
<details>
<summary>TPC-H: <b>Total hot run time: 31870 ms</b></summary>
```
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/tpch-tools
Tpch sf100 test result on commit 43e1e44a64ab2448408c6da658809b9120ff5ffb, data reload: false
------ Round 1 ----------------------------------
q1 17672 4289 4155 4155
q2 2075 382 239 239
q3 10153 1306 732 732
q4 10199 829 324 324
q5 7505 2040 1901 1901
q6 187 172 143 143
q7 946 788 658 658
q8 9278 1427 1217 1217
q9 4898 4786 4559 4559
q10 6714 1780 1425 1425
q11 487 306 280 280
q12 725 727 586 586
q13 17762 3868 3144 3144
q14 287 298 274 274
q15 592 522 507 507
q16 700 682 655 655
q17 662 766 525 525
q18 6750 6467 6473 6467
q19 1097 987 613 613
q20 396 358 249 249
q21 3023 2425 2264 2264
q22 1017 953 962 953
Total cold run time: 103125 ms
Total hot run time: 31870 ms
----- Round 2, with runtime_filter_mode=off -----
q1 4074 4079 4049 4049
q2 344 404 320 320
q3 2148 2640 2214 2214
q4 1339 1799 1330 1330
q5 4050 3981 3997 3981
q6 212 172 130 130
q7 1917 1819 1677 1677
q8 2950 2505 2548 2505
q9 7323 7196 7198 7196
q10 2564 2736 2278 2278
q11 560 479 444 444
q12 685 790 610 610
q13 3685 4161 3462 3462
q14 291 314 304 304
q15 541 503 514 503
q16 639 671 647 647
q17 1189 1387 1384 1384
q18 7939 8076 7769 7769
q19 905 876 908 876
q20 1971 2054 1949 1949
q21 4786 4517 4537 4517
q22 1183 1105 968 968
Total cold run time: 51295 ms
Total hot run time: 49113 ms
```
</details>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-3759561426
<details>
<summary>ClickBench: <b>Total hot run time: 26.92 s</b></summary>
```
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/clickbench-tools
ClickBench test result on commit 43e1e44a64ab2448408c6da658809b9120ff5ffb, data reload: false
query1 0.05 0.05 0.04
query2 0.10 0.04 0.04
query3 0.26 0.08 0.08
query4 1.61 0.12 0.11
query5 0.27 0.24 0.25
query6 1.14 0.67 0.64
query7 0.03 0.03 0.03
query8 0.05 0.04 0.04
query9 0.55 0.51 0.49
query10 0.55 0.54 0.54
query11 0.14 0.09 0.10
query12 0.14 0.11 0.11
query13 0.60 0.58 0.59
query14 0.95 0.94 0.95
query15 0.79 0.77 0.79
query16 0.40 0.40 0.38
query17 1.04 1.07 1.06
query18 0.23 0.21 0.22
query19 1.97 1.93 1.90
query20 0.02 0.01 0.01
query21 15.43 0.27 0.14
query22 5.29 0.05 0.04
query23 16.04 0.28 0.11
query24 1.72 0.52 0.30
query25 0.08 0.06 0.06
query26 0.14 0.13 0.14
query27 0.08 0.07 0.04
query28 3.79 1.05 0.88
query29 12.56 3.86 3.14
query30 0.27 0.13 0.12
query31 2.82 0.64 0.40
query32 3.24 0.56 0.47
query33 3.02 3.06 3.00
query34 16.30 5.11 4.39
query35 4.48 4.48 5.17
query36 0.71 0.56 0.60
query37 0.11 0.07 0.06
query38 0.07 0.05 0.04
query39 0.05 0.03 0.03
query40 0.17 0.15 0.14
query41 0.09 0.04 0.03
query42 0.04 0.03 0.03
query43 0.04 0.03 0.04
Total cold run time: 97.43 s
Total hot run time: 26.92 s
```
</details>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-3759822420
# BE UT Coverage Report
Increment line coverage `0.00% (0/9)` :tada:
[Increment coverage report](http://coverage.selectdb-in.cc/coverage/43e1e44a64ab2448408c6da658809b9120ff5ffb_43e1e44a64ab2448408c6da658809b9120ff5ffb/increment_report/index.html)
[Complete coverage report](http://coverage.selectdb-in.cc/coverage/43e1e44a64ab2448408c6da658809b9120ff5ffb_43e1e44a64ab2448408c6da658809b9120ff5ffb/report/index.html)
| Category | Coverage |
|-------------------|--------------------|
| Function Coverage | 52.39% (19029/36319) |
| Line Coverage | 35.71% (176328/493723) |
| Region Coverage | 32.25% (136520/423344) |
| Branch Coverage | 33.14% (59014/178061) |
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-3759537657
<details>
<summary>TPC-DS: <b>Total hot run time: 173685 ms</b></summary>
```
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/tpcds-tools
TPC-DS sf100 test result on commit 43e1e44a64ab2448408c6da658809b9120ff5ffb, data reload: false
query5 4427 628 483 483
query6 345 223 212 212
query7 4225 477 261 261
query8 329 240 239 239
query9 8658 2880 2894 2880
query10 508 376 309 309
query11 15286 15184 14847 14847
query12 171 115 110 110
query13 1247 472 369 369
query14 6240 3120 2784 2784
query14_1 2664 2614 2648 2614
query15 193 192 175 175
query16 975 415 472 415
query17 1074 643 537 537
query18 2472 455 321 321
query19 223 213 207 207
query20 123 114 116 114
query21 212 139 118 118
query22 3859 3900 3818 3818
query23 16019 15729 15340 15340
query23_1 15257 15570 15405 15405
query24 7170 1533 1188 1188
query24_1 1163 1199 1178 1178
query25 510 440 389 389
query26 1242 262 150 150
query27 2776 452 282 282
query28 4564 2166 2146 2146
query29 762 526 429 429
query30 319 236 213 213
query31 800 632 563 563
query32 93 101 76 76
query33 516 350 306 306
query34 914 866 525 525
query35 725 745 676 676
query36 883 937 856 856
query37 141 104 83 83
query38 2723 2757 2630 2630
query39 779 748 709 709
query39_1 708 702 716 702
query40 221 142 118 118
query41 67 63 63 63
query42 103 100 98 98
query43 470 422 432 422
query44 1345 757 748 748
query45 188 197 178 178
query46 830 945 586 586
query47 1407 1453 1383 1383
query48 310 333 241 241
query49 612 416 336 336
query50 625 269 204 204
query51 3785 3873 3843 3843
query52 105 110 98 98
query53 285 323 278 278
query54 311 263 256 256
query55 79 85 81 81
query56 303 296 295 295
query57 1024 1014 908 908
query58 267 267 251 251
query59 2129 2168 2030 2030
query60 356 346 337 337
query61 175 171 167 167
query62 397 356 312 312
query63 301 266 264 264
query64 5102 1362 1045 1045
query65 3737 3704 3759 3704
query66 1483 429 331 331
query67 15629 15687 15405 15405
query68 2387 1120 774 774
query69 450 361 333 333
query70 981 969 880 880
query71 324 316 297 297
query72 5486 3347 3313 3313
query73 620 718 304 304
query74 8752 8701 8548 8548
query75 2752 2821 2466 2466
query76 2290 1056 663 663
query77 373 384 306 306
query78 9783 9906 9136 9136
query79 1067 912 599 599
query80 1312 577 466 466
query81 552 258 237 237
query82 1164 140 109 109
query83 375 259 238 238
query84 252 112 98 98
query85 886 478 423 423
query86 416 326 290 290
query87 2850 2866 2741 2741
query88 3513 2611 2539 2539
query89 390 357 338 338
query90 1939 176 167 167
query91 169 157 134 134
query92 76 70 71 70
query93 945 930 536 536
query94 646 318 301 301
query95 597 396 317 317
query96 634 501 234 234
query97 2376 2363 2301 2301
query98 207 200 196 196
query99 586 575 518 518
Total cold run time: 247049 ms
Total hot run time: 173685 ms
```
</details>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "Thearas (via GitHub)" <gi...@apache.org>.
Thearas commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-3758778619
Thank you for your contribution to Apache Doris.
Don't know what should be done next? See [How to process your PR](https://cwiki.apache.org/confluence/display/DORIS/How+to+process+your+PR).
Please clearly describe your PR:
1. What problem was fixed (it's best to include specific error reporting information). How it was fixed.
2. Which behaviors were modified. What was the previous behavior, what is it now, why was it modified, and what possible impacts might there be.
3. What features were added. Why was this function added?
4. Which code was refactored and why was this part of the code refactored?
5. Which functions were optimized and what is the difference before and after the optimization?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org
Re: [PR] [opt](ann index) Introduce ann_topn_min_input_rows_ratio [doris]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #59968:
URL: https://github.com/apache/doris/pull/59968#issuecomment-5008787578
We're closing this PR because it hasn't been updated in a while.
This isn't a judgement on the merit of the PR in any way. It's just a way of keeping the PR queue manageable.
If you'd like to revive this PR, please reopen it and feel free a maintainer to remove the Stale tag!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org