You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@doris.apache.org by "Gabriel39 (via GitHub)" <gi...@apache.org> on 2025/11/17 03:33:04 UTC

[PR] [Improvement](external) Increase scanner concurrency [doris]

Gabriel39 opened a new pull request, #58073:
URL: https://github.com/apache/doris/pull/58073

   ### What problem does this PR solve?
   
   The concurrency of scanners are controlled by some variables which is not suitable for different cases. For external tables, we need more scanners than the internal tables. In this PR, the concurrency variable is splitted from the origin one. 
   
   ### 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2562531645


##########
be/src/pipeline/exec/file_scan_operator.cpp:
##########
@@ -31,6 +31,39 @@
 
 namespace doris::pipeline {
 #include "common/compile_check_begin.h"
+
+int FileScanLocalState::max_scanners_concurrency(RuntimeState* state) const {
+    // For select * from table limit 10; should just use one thread.
+    if (should_run_serial()) {
+        return 1;
+    }
+    /*
+     * The max concurrency of file scanners for each FileScanLocalState is determined by:
+     * 1. User specified max_file_scanners_concurrency which is set through session variable.
+     * 2. Default: std::max(512, CpuInfo::num_cores() * 10)
+     *
+     * If this is a serial operator, the max concurrency should multiply by the number of parallel instances of the operator.
+     */
+    return (state->max_file_scanners_concurrency() ? state->max_file_scanners_concurrency()
+                                                   : std::max(512, CpuInfo::num_cores() * 10) * 2) *
+           (state->query_parallel_instance_num() / _parent->parallelism(state));
+}
+
+int FileScanLocalState::min_scanners_concurrency(RuntimeState* state) const {
+    /*
+     * The min concurrency of scanners for each ScanLocalStateBase is determined by:
+     * 1. User specified min_scan_concurrency_of_scan_scheduler which is set through session variable.
+     * 2. Default: 2 * state->get_query_ctx()->get_remote_scan_scheduler()->get_max_threads()
+     */
+    return state->min_scan_concurrency_of_scan_scheduler()
+                   ? state->min_scan_concurrency_of_scan_scheduler()
+                   : 2 * state->get_query_ctx()->get_remote_scan_scheduler()->get_max_threads();

Review Comment:
   这里换成跟 cpu core的关系,不要依赖max threads



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3581702724

   # FE Regression Coverage Report
   Increment line coverage `  80.00% (8/10)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0_merge_fe/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0_merge_fe/report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584445244

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3540124012

   
   <details>
   <summary>ClickBench: <b>Total hot run time: 28.52 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 2e0bd88b8407116b6be9c54135c09920f79b3756, data reload: false
   
   query1	0.05	0.04	0.04
   query2	0.14	0.07	0.07
   query3	0.32	0.08	0.08
   query4	1.61	0.10	0.09
   query5	0.27	0.27	0.25
   query6	1.19	0.64	0.64
   query7	0.03	0.03	0.02
   query8	0.08	0.06	0.06
   query9	0.60	0.53	0.52
   query10	0.58	0.58	0.56
   query11	0.25	0.14	0.14
   query12	0.26	0.15	0.15
   query13	0.63	0.64	0.61
   query14	1.01	1.01	1.00
   query15	0.86	0.82	0.82
   query16	0.39	0.41	0.40
   query17	1.08	1.06	0.98
   query18	0.24	0.23	0.22
   query19	1.92	1.91	1.84
   query20	0.02	0.02	0.01
   query21	15.42	0.29	0.24
   query22	5.00	0.12	0.10
   query23	15.39	0.36	0.23
   query24	2.28	0.45	0.32
   query25	0.10	0.10	0.10
   query26	0.20	0.18	0.18
   query27	0.10	0.09	0.09
   query28	3.70	1.26	1.10
   query29	12.55	4.03	3.34
   query30	0.35	0.13	0.12
   query31	2.82	0.64	0.44
   query32	3.28	0.60	0.51
   query33	3.04	3.06	3.08
   query34	16.09	5.14	4.59
   query35	4.59	4.61	4.61
   query36	0.63	0.51	0.49
   query37	0.24	0.09	0.09
   query38	0.20	0.05	0.06
   query39	0.06	0.04	0.05
   query40	0.22	0.17	0.16
   query41	0.12	0.08	0.06
   query42	0.07	0.05	0.05
   query43	0.07	0.05	0.06
   Total cold run time: 98.05 s
   Total hot run time: 28.52 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3570334601

   
   <details>
   <summary>TPC-H: <b>Total hot run time: 34278 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 6d4a5a626312a4f4c9a77a0ad5c412dca03620b0, data reload: false
   
   ------ Round 1 ----------------------------------
   q1	17594	5044	4891	4891
   q2	2029	314	199	199
   q3	10207	1295	713	713
   q4	10239	960	364	364
   q5	7554	2531	2263	2263
   q6	190	173	139	139
   q7	905	773	629	629
   q8	9290	1336	1020	1020
   q9	7329	5357	5377	5357
   q10	6861	2231	1825	1825
   q11	500	306	294	294
   q12	359	370	228	228
   q13	17796	3739	3050	3050
   q14	243	241	218	218
   q15	572	511	527	511
   q16	1030	1011	934	934
   q17	593	747	498	498
   q18	7378	7181	7144	7144
   q19	1102	953	563	563
   q20	345	351	228	228
   q21	3582	2605	2292	2292
   q22	1012	968	918	918
   Total cold run time: 106710 ms
   Total hot run time: 34278 ms
   
   ----- Round 2, with runtime_filter_mode=off -----
   q1	4943	4892	4904	4892
   q2	332	393	320	320
   q3	2133	2695	2298	2298
   q4	1305	1758	1293	1293
   q5	4187	4314	4422	4314
   q6	227	173	135	135
   q7	2048	1940	1836	1836
   q8	2664	2444	2463	2444
   q9	7625	7648	7590	7590
   q10	3059	3250	2830	2830
   q11	586	515	504	504
   q12	693	747	652	652
   q13	3720	3849	3352	3352
   q14	288	327	289	289
   q15	528	510	512	510
   q16	1062	1086	1036	1036
   q17	1144	1506	1416	1416
   q18	7819	7864	7513	7513
   q19	766	749	761	749
   q20	2157	2039	1911	1911
   q21	4769	4249	4193	4193
   q22	1051	988	967	967
   Total cold run time: 53106 ms
   Total hot run time: 51044 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3570695566

   # BE UT Coverage Report
   Increment line coverage `18.60% (8/43)` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/6d4a5a626312a4f4c9a77a0ad5c412dca03620b0_6d4a5a626312a4f4c9a77a0ad5c412dca03620b0/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/6d4a5a626312a4f4c9a77a0ad5c412dca03620b0_6d4a5a626312a4f4c9a77a0ad5c412dca03620b0/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 52.63% (18280/34735) |
   | Line Coverage     | 38.05% (166262/436947) |
   | Region Coverage   | 33.00% (129229/391573) |
   | Branch Coverage   | 33.80% (55461/164082) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3542713118

   # BE UT Coverage Report
   Increment line coverage `20.51% (8/39)` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/22249e2f89b2266965806f7444efdeb3e16ab171_22249e2f89b2266965806f7444efdeb3e16ab171/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/22249e2f89b2266965806f7444efdeb3e16ab171_22249e2f89b2266965806f7444efdeb3e16ab171/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 52.74% (18274/34646) |
   | Line Coverage     | 38.12% (166087/435738) |
   | Region Coverage   | 33.08% (129138/390377) |
   | Branch Coverage   | 33.83% (55395/163759) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3569682292

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "morningman (via GitHub)" <gi...@apache.org>.
morningman commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2555033634


##########
be/src/pipeline/exec/file_scan_operator.cpp:
##########
@@ -31,6 +31,15 @@
 
 namespace doris::pipeline {
 #include "common/compile_check_begin.h"
+
+int FileScanLocalState::max_scanners_concurrency(RuntimeState* state) {
+    return (state->max_file_scanners_concurrency()

Review Comment:
   Add some comment for this logic



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2562533157


##########
be/src/pipeline/exec/file_scan_operator.cpp:
##########
@@ -31,6 +31,39 @@
 
 namespace doris::pipeline {
 #include "common/compile_check_begin.h"
+
+int FileScanLocalState::max_scanners_concurrency(RuntimeState* state) const {
+    // For select * from table limit 10; should just use one thread.
+    if (should_run_serial()) {
+        return 1;
+    }
+    /*
+     * The max concurrency of file scanners for each FileScanLocalState is determined by:
+     * 1. User specified max_file_scanners_concurrency which is set through session variable.
+     * 2. Default: std::max(512, CpuInfo::num_cores() * 10)
+     *
+     * If this is a serial operator, the max concurrency should multiply by the number of parallel instances of the operator.
+     */
+    return (state->max_file_scanners_concurrency() ? state->max_file_scanners_concurrency()

Review Comment:
   max_file_scanners_concurrency > 0



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3578475777

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2562529484


##########
be/src/pipeline/exec/file_scan_operator.cpp:
##########
@@ -31,6 +31,39 @@
 
 namespace doris::pipeline {
 #include "common/compile_check_begin.h"
+
+int FileScanLocalState::max_scanners_concurrency(RuntimeState* state) const {
+    // For select * from table limit 10; should just use one thread.
+    if (should_run_serial()) {
+        return 1;
+    }
+    /*
+     * The max concurrency of file scanners for each FileScanLocalState is determined by:
+     * 1. User specified max_file_scanners_concurrency which is set through session variable.
+     * 2. Default: std::max(512, CpuInfo::num_cores() * 10)
+     *
+     * If this is a serial operator, the max concurrency should multiply by the number of parallel instances of the operator.
+     */
+    return (state->max_file_scanners_concurrency() ? state->max_file_scanners_concurrency()
+                                                   : std::max(512, CpuInfo::num_cores() * 10) * 2) *
+           (state->query_parallel_instance_num() / _parent->parallelism(state));
+}
+
+int FileScanLocalState::min_scanners_concurrency(RuntimeState* state) const {
+    /*
+     * The min concurrency of scanners for each ScanLocalStateBase is determined by:
+     * 1. User specified min_scan_concurrency_of_scan_scheduler which is set through session variable.
+     * 2. Default: 2 * state->get_query_ctx()->get_remote_scan_scheduler()->get_max_threads()
+     */
+    return state->min_scan_concurrency_of_scan_scheduler()

Review Comment:
   min_scan_concurrency_of_scan_scheduler > 0



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3578602796

   # Cloud UT Coverage Report
   Increment line coverage ` ` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/98f7f81cbd7268940b4d1428e678607919e50b8c_98f7f81cbd7268940b4d1428e678607919e50b8c_cloud/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/98f7f81cbd7268940b4d1428e678607919e50b8c_98f7f81cbd7268940b4d1428e678607919e50b8c_cloud/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 80.86% (1707/2111) |
   | Line Coverage     | 66.85% (29889/44708) |
   | Region Coverage   | 67.38% (14900/22114) |
   | Branch Coverage   | 57.56% (7925/13768) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3575078848

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3580676140

   # Cloud UT Coverage Report
   Increment line coverage ` ` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0_dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0_cloud/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0_dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0_cloud/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 80.86% (1707/2111) |
   | Line Coverage     | 66.84% (29881/44708) |
   | Region Coverage   | 67.38% (14900/22114) |
   | Branch Coverage   | 57.58% (7927/13768) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3539931389

   # Cloud UT Coverage Report
   Increment line coverage ` ` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/2e0bd88b8407116b6be9c54135c09920f79b3756_2e0bd88b8407116b6be9c54135c09920f79b3756_cloud/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/2e0bd88b8407116b6be9c54135c09920f79b3756_2e0bd88b8407116b6be9c54135c09920f79b3756_cloud/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 80.56% (1683/2089) |
   | Line Coverage     | 66.71% (29696/44518) |
   | Region Coverage   | 67.27% (14819/22028) |
   | Branch Coverage   | 57.54% (7886/13706) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3542288369

   
   <details>
   <summary>TPC-H: <b>Total hot run time: 36006 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 22249e2f89b2266965806f7444efdeb3e16ab171, data reload: false
   
   ------ Round 1 ----------------------------------
   q1	17401	5569	5188	5188
   q2	2010	307	206	206
   q3	10161	1242	713	713
   q4	10225	897	348	348
   q5	7498	2443	2354	2354
   q6	187	176	137	137
   q7	939	761	625	625
   q8	9355	1426	1319	1319
   q9	7375	5294	5418	5294
   q10	6912	2263	1789	1789
   q11	501	292	272	272
   q12	331	367	220	220
   q13	17784	3627	3016	3016
   q14	241	236	208	208
   q15	569	529	499	499
   q16	987	1026	928	928
   q17	593	873	342	342
   q18	8234	7529	7665	7529
   q19	1123	977	597	597
   q20	380	369	256	256
   q21	4242	3384	3099	3099
   q22	1200	1081	1067	1067
   Total cold run time: 108248 ms
   Total hot run time: 36006 ms
   
   ----- Round 2, with runtime_filter_mode=off -----
   q1	5489	5571	5486	5486
   q2	245	326	230	230
   q3	2489	2810	2467	2467
   q4	1478	1879	1461	1461
   q5	4571	4948	5022	4948
   q6	177	160	126	126
   q7	2014	1963	2009	1963
   q8	2543	2828	2623	2623
   q9	7402	7527	7230	7230
   q10	2874	3167	2623	2623
   q11	573	504	497	497
   q12	625	709	574	574
   q13	3267	3651	3001	3001
   q14	272	273	253	253
   q15	526	503	477	477
   q16	1021	1043	979	979
   q17	1180	1694	1219	1219
   q18	7288	7261	7141	7141
   q19	760	988	1060	988
   q20	1909	1940	1823	1823
   q21	4758	4399	4297	4297
   q22	1098	1039	995	995
   Total cold run time: 52559 ms
   Total hot run time: 51401 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2562517054


##########
fe/fe-core/src/main/java/org/apache/doris/qe/SessionVariable.java:
##########
@@ -977,12 +978,17 @@ public static double getHotValueThreshold() {
     // 100MB
     public long maxScanQueueMemByte = 2147483648L / 20;
 
-    @VariableMgr.VarAttr(name = NUM_SCANNER_THREADS, needForward = true, description = {
+    @VariableMgr.VarAttr(name = MAX_SCANNERS_CONCURRENCY, needForward = true, description = {

Review Comment:
   max_local_scanners_concurrency   vs max_remote_scanners_concurrency ?



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3581070922

   
   <details>
   <summary>TPC-DS: <b>Total hot run time: 184376 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 dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0, data reload: false
   
   query1	1058	406	389	389
   query2	6578	1572	1566	1566
   query3	6755	224	218	218
   query4	25300	23225	22653	22653
   query5	4963	622	464	464
   query6	341	238	242	238
   query7	4648	513	302	302
   query8	325	259	241	241
   query9	8728	2556	2576	2556
   query10	539	334	301	301
   query11	15232	15050	14994	14994
   query12	191	123	117	117
   query13	1706	587	462	462
   query14	11121	8858	8747	8747
   query15	214	199	187	187
   query16	7688	719	526	526
   query17	1212	774	651	651
   query18	2048	435	346	346
   query19	214	216	196	196
   query20	132	127	120	120
   query21	217	141	115	115
   query22	3776	3957	3836	3836
   query23	32835	31860	32029	31860
   query24	8473	2431	2408	2408
   query25	653	562	495	495
   query26	1249	280	172	172
   query27	2714	496	355	355
   query28	4308	2127	2098	2098
   query29	837	658	518	518
   query30	311	238	209	209
   query31	845	706	637	637
   query32	92	82	76	76
   query33	612	396	355	355
   query34	805	915	550	550
   query35	811	824	739	739
   query36	887	913	833	833
   query37	130	117	95	95
   query38	3350	3361	3326	3326
   query39	1480	1454	1458	1454
   query40	227	129	126	126
   query41	67	60	61	60
   query42	123	111	108	108
   query43	437	443	423	423
   query44	1277	765	755	755
   query45	201	191	188	188
   query46	875	991	635	635
   query47	1704	1738	1637	1637
   query48	399	415	329	329
   query49	774	495	411	411
   query50	645	674	402	402
   query51	3881	3905	3868	3868
   query52	111	108	110	108
   query53	232	259	208	208
   query54	305	290	279	279
   query55	97	94	91	91
   query56	338	339	344	339
   query57	1140	1170	1115	1115
   query58	292	269	273	269
   query59	2320	2445	2382	2382
   query60	349	350	349	349
   query61	160	161	160	160
   query62	769	700	663	663
   query63	230	201	193	193
   query64	4512	1230	900	900
   query65	4088	3923	3938	3923
   query66	1098	441	348	348
   query67	15467	15069	14932	14932
   query68	8317	961	629	629
   query69	515	354	306	306
   query70	1280	1222	1228	1222
   query71	490	329	314	314
   query72	5837	4773	4844	4773
   query73	674	577	341	341
   query74	8825	8805	8565	8565
   query75	3978	3307	2792	2792
   query76	3815	1115	690	690
   query77	808	402	308	308
   query78	9463	9651	8872	8872
   query79	2046	842	597	597
   query80	643	600	493	493
   query81	492	272	233	233
   query82	411	163	136	136
   query83	271	259	252	252
   query84	267	115	106	106
   query85	914	484	445	445
   query86	349	318	300	300
   query87	3466	3443	3412	3412
   query88	3794	2252	2269	2252
   query89	389	327	298	298
   query90	1983	220	209	209
   query91	174	170	143	143
   query92	85	75	65	65
   query93	1414	991	659	659
   query94	726	471	346	346
   query95	499	416	403	403
   query96	505	547	289	289
   query97	2913	2938	2843	2843
   query98	237	211	213	211
   query99	1394	1376	1266	1266
   Total cold run time: 272470 ms
   Total hot run time: 184376 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2562540277


##########
be/src/pipeline/exec/scan_operator.cpp:
##########
@@ -72,6 +72,38 @@ bool ScanLocalState<Derived>::should_run_serial() const {
     return _parent->cast<typename Derived::Parent>()._should_run_serial;
 }
 
+int ScanLocalStateBase::max_scanners_concurrency(RuntimeState* state) const {
+    // For select * from table limit 10; should just use one thread.
+    if (should_run_serial()) {
+        return 1;
+    }
+    /*
+     * The max concurrency of scanners for each ScanLocalStateBase is determined by:
+     * 1. User specified max_scanners_concurrency which is set through session variable.
+     * 2. Default: std::max(48, CpuInfo::num_cores() * 2))
+     *
+     * If this is a serial operator, the max concurrency should multiply by the number of parallel instances of the operator.
+     */
+    return (state->max_scanners_concurrency() ? state->max_scanners_concurrency()
+                                              : std::max(48, CpuInfo::num_cores() * 2) * 2) *
+           (state->query_parallel_instance_num() / _parent->parallelism(state));
+}
+
+int ScanLocalStateBase::min_scanners_concurrency(RuntimeState* state) const {
+    /*
+     * The min concurrency of scanners for each ScanLocalStateBase is determined by:
+     * 1. User specified min_scan_concurrency_of_scan_scheduler which is set through session variable.
+     * 2. Default: 2 * state->get_query_ctx()->get_scan_scheduler()->get_max_threads()
+     */
+    return state->min_scan_concurrency_of_scan_scheduler()
+                   ? state->min_scan_concurrency_of_scan_scheduler()
+                   : 2 * state->get_query_ctx()->get_scan_scheduler()->get_max_threads();

Review Comment:
   跟cpu 关联,不要跟threads 关联



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Thearas (via GitHub)" <gi...@apache.org>.
Thearas commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3539815989

   
   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584264916

   
   <details>
   <summary>TPC-DS: <b>Total hot run time: 184571 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 b87eaf23bd13dcfd21756e0823e75065d36d475b, data reload: false
   
   query1	1110	427	392	392
   query2	6578	1577	1615	1577
   query3	6750	222	226	222
   query4	25674	23362	22880	22880
   query5	4902	625	483	483
   query6	343	249	226	226
   query7	4642	510	309	309
   query8	302	252	240	240
   query9	8708	2577	2572	2572
   query10	545	373	302	302
   query11	15159	15203	14977	14977
   query12	180	117	115	115
   query13	1707	571	484	484
   query14	11073	8907	8749	8749
   query15	212	203	194	194
   query16	7526	703	541	541
   query17	1220	793	655	655
   query18	2013	416	322	322
   query19	207	198	171	171
   query20	133	124	119	119
   query21	215	147	115	115
   query22	3996	3951	3877	3877
   query23	33039	31988	31758	31758
   query24	8410	2383	2387	2383
   query25	604	518	444	444
   query26	1243	283	168	168
   query27	2736	487	339	339
   query28	4339	2133	2086	2086
   query29	831	639	491	491
   query30	314	240	207	207
   query31	852	732	589	589
   query32	81	74	73	73
   query33	600	377	335	335
   query34	800	874	546	546
   query35	802	855	763	763
   query36	896	942	851	851
   query37	122	111	92	92
   query38	3356	3326	3284	3284
   query39	1461	1384	1403	1384
   query40	234	132	117	117
   query41	65	64	63	63
   query42	124	118	110	110
   query43	429	460	420	420
   query44	1293	759	755	755
   query45	201	188	184	184
   query46	874	997	645	645
   query47	1692	1703	1631	1631
   query48	400	423	328	328
   query49	781	503	428	428
   query50	659	677	410	410
   query51	3885	3926	3838	3838
   query52	112	113	99	99
   query53	243	263	189	189
   query54	309	291	278	278
   query55	98	96	89	89
   query56	336	335	331	331
   query57	1156	1148	1090	1090
   query58	295	275	269	269
   query59	2389	2364	2236	2236
   query60	346	344	339	339
   query61	166	164	167	164
   query62	799	707	681	681
   query63	227	195	195	195
   query64	4561	1306	1014	1014
   query65	4085	3962	4023	3962
   query66	1142	466	363	363
   query67	15486	14995	14845	14845
   query68	7399	952	628	628
   query69	525	369	315	315
   query70	1287	1203	1210	1203
   query71	442	347	328	328
   query72	6036	4863	4815	4815
   query73	659	573	341	341
   query74	8815	8753	8683	8683
   query75	3329	3342	2865	2865
   query76	3329	1126	699	699
   query77	509	405	330	330
   query78	9476	9642	8940	8940
   query79	2276	803	593	593
   query80	634	580	531	531
   query81	520	272	249	249
   query82	431	160	131	131
   query83	260	262	244	244
   query84	256	107	97	97
   query85	924	487	452	452
   query86	382	296	282	282
   query87	3485	3474	3463	3463
   query88	4125	2237	2250	2237
   query89	397	330	297	297
   query90	1931	222	216	216
   query91	174	171	139	139
   query92	89	74	69	69
   query93	1975	987	663	663
   query94	790	470	366	366
   query95	496	404	408	404
   query96	509	565	300	300
   query97	2893	2981	2842	2842
   query98	244	221	210	210
   query99	1352	1419	1251	1251
   Total cold run time: 272017 ms
   Total hot run time: 184571 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584057703

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3540109421

   
   <details>
   <summary>TPC-DS: <b>Total hot run time: 190710 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 2e0bd88b8407116b6be9c54135c09920f79b3756, data reload: false
   
   query1	1011	400	422	400
   query2	6573	1761	1739	1739
   query3	6758	241	229	229
   query4	26544	23654	23467	23467
   query5	4456	660	551	551
   query6	331	256	238	238
   query7	4657	489	297	297
   query8	301	264	254	254
   query9	8704	2946	2868	2868
   query10	513	376	314	314
   query11	15546	15061	14854	14854
   query12	191	124	121	121
   query13	1697	592	533	533
   query14	10560	9604	9639	9604
   query15	243	188	178	178
   query16	7903	693	500	500
   query17	1660	844	645	645
   query18	2053	436	343	343
   query19	250	220	200	200
   query20	137	133	125	125
   query21	216	139	116	116
   query22	3963	4123	4126	4123
   query23	34124	33303	32748	32748
   query24	8083	2426	2403	2403
   query25	623	575	490	490
   query26	1230	279	177	177
   query27	2690	501	374	374
   query28	4362	2338	2309	2309
   query29	826	673	544	544
   query30	306	225	200	200
   query31	887	831	769	769
   query32	95	89	84	84
   query33	589	409	370	370
   query34	773	862	572	572
   query35	850	883	769	769
   query36	988	1013	909	909
   query37	134	123	94	94
   query38	3602	3581	3594	3581
   query39	1501	1439	1414	1414
   query40	239	138	148	138
   query41	65	59	61	59
   query42	140	135	123	123
   query43	492	501	458	458
   query44	1292	806	814	806
   query45	191	184	178	178
   query46	879	991	649	649
   query47	1759	1787	1713	1713
   query48	413	424	338	338
   query49	772	487	437	437
   query50	662	689	418	418
   query51	3952	4022	3930	3930
   query52	127	123	117	117
   query53	246	274	201	201
   query54	330	315	324	315
   query55	92	96	92	92
   query56	347	370	359	359
   query57	1172	1178	1125	1125
   query58	322	297	342	297
   query59	2898	2960	2925	2925
   query60	374	382	351	351
   query61	164	159	152	152
   query62	764	704	660	660
   query63	249	214	200	200
   query64	4443	1164	883	883
   query65	4079	3967	3953	3953
   query66	1081	442	353	353
   query67	15596	15305	14876	14876
   query68	8296	982	660	660
   query69	488	349	311	311
   query70	1268	1179	1204	1179
   query71	478	376	350	350
   query72	6002	5144	5383	5144
   query73	676	750	383	383
   query74	9068	8836	8748	8748
   query75	3904	3288	2820	2820
   query76	3749	1172	791	791
   query77	821	448	346	346
   query78	9706	9743	8980	8980
   query79	3014	854	638	638
   query80	760	611	536	536
   query81	513	270	227	227
   query82	373	171	137	137
   query83	301	268	250	250
   query84	302	115	102	102
   query85	917	479	439	439
   query86	390	296	284	284
   query87	3786	3762	3693	3693
   query88	3608	2313	2284	2284
   query89	400	340	302	302
   query90	1899	255	235	235
   query91	183	164	137	137
   query92	80	77	72	72
   query93	2078	1001	685	685
   query94	688	436	333	333
   query95	404	335	344	335
   query96	495	593	289	289
   query97	2971	2969	2898	2898
   query98	266	225	230	225
   query99	1505	1411	1285	1285
   Total cold run time: 278459 ms
   Total hot run time: 190710 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3542351576

   
   <details>
   <summary>TPC-DS: <b>Total hot run time: 189875 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 22249e2f89b2266965806f7444efdeb3e16ab171, data reload: false
   
   query1	1054	406	426	406
   query2	6543	1727	1749	1727
   query3	6755	227	220	220
   query4	26577	23441	23226	23226
   query5	4410	658	531	531
   query6	330	243	234	234
   query7	4647	485	299	299
   query8	305	269	242	242
   query9	8674	2910	2878	2878
   query10	517	363	317	317
   query11	15621	15136	14834	14834
   query12	181	125	116	116
   query13	1664	580	462	462
   query14	10564	9108	9167	9108
   query15	198	185	173	173
   query16	7797	640	505	505
   query17	1376	742	601	601
   query18	2039	431	315	315
   query19	244	222	237	222
   query20	135	133	123	123
   query21	211	130	118	118
   query22	4186	4422	4280	4280
   query23	34118	32974	32894	32894
   query24	8050	2261	2379	2261
   query25	588	532	465	465
   query26	1241	279	160	160
   query27	2720	496	361	361
   query28	4414	2303	2279	2279
   query29	786	626	537	537
   query30	301	227	199	199
   query31	881	787	711	711
   query32	94	80	80	80
   query33	566	384	358	358
   query34	763	851	527	527
   query35	830	858	756	756
   query36	983	1023	881	881
   query37	124	119	99	99
   query38	3582	3486	3507	3486
   query39	1472	1421	1417	1417
   query40	230	135	124	124
   query41	64	61	60	60
   query42	138	121	118	118
   query43	474	496	447	447
   query44	1281	806	799	799
   query45	192	195	185	185
   query46	880	966	661	661
   query47	1804	1853	1776	1776
   query48	413	431	355	355
   query49	793	530	450	450
   query50	655	676	411	411
   query51	3890	4020	3912	3912
   query52	124	119	110	110
   query53	237	276	207	207
   query54	324	312	304	304
   query55	91	94	88	88
   query56	337	351	378	351
   query57	1169	1193	1126	1126
   query58	315	305	304	304
   query59	2863	2957	2909	2909
   query60	394	393	363	363
   query61	193	185	195	185
   query62	766	740	687	687
   query63	236	202	199	199
   query64	4592	1314	988	988
   query65	4033	3917	3943	3917
   query66	1093	455	373	373
   query67	15446	15152	15088	15088
   query68	8114	949	662	662
   query69	509	347	314	314
   query70	1277	1236	1229	1229
   query71	440	378	356	356
   query72	6395	5123	5315	5123
   query73	657	760	370	370
   query74	8899	9124	8779	8779
   query75	3299	3262	2857	2857
   query76	3288	1119	748	748
   query77	548	411	349	349
   query78	9755	9852	8978	8978
   query79	2845	820	612	612
   query80	680	602	526	526
   query81	503	263	228	228
   query82	438	169	139	139
   query83	298	271	248	248
   query84	306	111	101	101
   query85	927	484	444	444
   query86	378	303	294	294
   query87	3694	3702	3622	3622
   query88	3674	2278	2275	2275
   query89	396	345	303	303
   query90	2000	236	223	223
   query91	169	164	137	137
   query92	85	78	71	71
   query93	2101	954	684	684
   query94	726	432	327	327
   query95	403	345	332	332
   query96	482	593	295	295
   query97	2941	2953	2854	2854
   query98	261	220	218	218
   query99	1446	1385	1272	1272
   Total cold run time: 276502 ms
   Total hot run time: 189875 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3569868773

   # Cloud UT Coverage Report
   Increment line coverage ` ` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/6d4a5a626312a4f4c9a77a0ad5c412dca03620b0_6d4a5a626312a4f4c9a77a0ad5c412dca03620b0_cloud/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/6d4a5a626312a4f4c9a77a0ad5c412dca03620b0_6d4a5a626312a4f4c9a77a0ad5c412dca03620b0_cloud/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 80.86% (1707/2111) |
   | Line Coverage     | 66.85% (29887/44708) |
   | Region Coverage   | 67.38% (14900/22114) |
   | Branch Coverage   | 57.57% (7926/13768) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584674096

   # FE Regression Coverage Report
   Increment line coverage `  0.00% (0/10)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_b87eaf23bd13dcfd21756e0823e75065d36d475b_merge_fe/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_b87eaf23bd13dcfd21756e0823e75065d36d475b_merge_fe/report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584901203

   skip check_coverage


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3579869432

   
   <details>
   <summary>TPC-H: <b>Total hot run time: 34726 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 2462a11bc42619b4fdcffd67b314243593f9dc64, data reload: false
   
   ------ Round 1 ----------------------------------
   q1	17610	5507	5115	5115
   q2	2041	315	215	215
   q3	10263	1243	729	729
   q4	10235	920	374	374
   q5	7467	2499	2100	2100
   q6	185	176	138	138
   q7	944	784	655	655
   q8	9339	1416	1230	1230
   q9	7319	5473	5550	5473
   q10	6883	2258	1782	1782
   q11	496	320	294	294
   q12	319	417	226	226
   q13	17785	3650	3128	3128
   q14	236	253	224	224
   q15	584	522	508	508
   q16	1039	1022	965	965
   q17	599	871	343	343
   q18	7691	7248	7193	7193
   q19	943	955	545	545
   q20	348	344	230	230
   q21	4000	3179	2315	2315
   q22	1069	998	944	944
   Total cold run time: 107395 ms
   Total hot run time: 34726 ms
   
   ----- Round 2, with runtime_filter_mode=off -----
   q1	5149	5178	5162	5162
   q2	316	401	316	316
   q3	2249	2687	2318	2318
   q4	1355	1838	1358	1358
   q5	4456	5277	5002	5002
   q6	183	170	135	135
   q7	2143	1997	1856	1856
   q8	2643	2808	2764	2764
   q9	8059	7520	7641	7520
   q10	3057	3354	2811	2811
   q11	618	525	504	504
   q12	774	798	616	616
   q13	3657	3967	3219	3219
   q14	302	324	323	323
   q15	581	513	510	510
   q16	1121	1168	1105	1105
   q17	1291	1791	1262	1262
   q18	8030	7760	7710	7710
   q19	841	916	1149	916
   q20	2041	2046	1975	1975
   q21	5126	4681	4489	4489
   q22	1113	1052	1000	1000
   Total cold run time: 55105 ms
   Total hot run time: 52871 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2563740017


##########
be/src/pipeline/exec/file_scan_operator.cpp:
##########
@@ -31,6 +31,39 @@
 
 namespace doris::pipeline {
 #include "common/compile_check_begin.h"
+
+int FileScanLocalState::max_scanners_concurrency(RuntimeState* state) const {
+    // For select * from table limit 10; should just use one thread.
+    if (should_run_serial()) {
+        return 1;
+    }
+    /*
+     * The max concurrency of file scanners for each FileScanLocalState is determined by:
+     * 1. User specified max_file_scanners_concurrency which is set through session variable.
+     * 2. Default: std::max(512, CpuInfo::num_cores() * 10)
+     *
+     * If this is a serial operator, the max concurrency should multiply by the number of parallel instances of the operator.
+     */
+    return (state->max_file_scanners_concurrency() ? state->max_file_scanners_concurrency()
+                                                   : std::max(512, CpuInfo::num_cores() * 10) * 2) *
+           (state->query_parallel_instance_num() / _parent->parallelism(state));

Review Comment:
   这里为什么要 乘 state->query_parallel_instance_num()



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3578778228

   # FE UT Coverage Report
   Increment line coverage `  66.67% (4/6)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_98f7f81cbd7268940b4d1428e678607919e50b8c/fe_increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_98f7f81cbd7268940b4d1428e678607919e50b8c/fe_report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584061178

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3570241802

   # FE UT Coverage Report
   Increment line coverage `  66.67% (4/6)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_6d4a5a626312a4f4c9a77a0ad5c412dca03620b0/fe_increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_6d4a5a626312a4f4c9a77a0ad5c412dca03620b0/fe_report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3578899786

   
   <details>
   <summary>TPC-H: <b>Total hot run time: 35292 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 98f7f81cbd7268940b4d1428e678607919e50b8c, data reload: false
   
   ------ Round 1 ----------------------------------
   q1	17621	5507	5097	5097
   q2	2019	330	220	220
   q3	10241	1239	733	733
   q4	10230	927	368	368
   q5	7499	2414	2134	2134
   q6	197	176	139	139
   q7	944	801	653	653
   q8	9378	1460	1339	1339
   q9	7507	5506	5513	5506
   q10	6898	2249	1794	1794
   q11	504	303	294	294
   q12	328	363	222	222
   q13	17807	3697	3039	3039
   q14	238	235	215	215
   q15	595	517	506	506
   q16	1030	1041	966	966
   q17	625	896	382	382
   q18	7654	7382	7042	7042
   q19	1103	948	548	548
   q20	354	352	230	230
   q21	3709	3113	2898	2898
   q22	1053	977	967	967
   Total cold run time: 107534 ms
   Total hot run time: 35292 ms
   
   ----- Round 2, with runtime_filter_mode=off -----
   q1	5144	5111	5135	5111
   q2	339	390	297	297
   q3	2278	2702	2316	2316
   q4	1330	1877	1367	1367
   q5	4470	5152	4973	4973
   q6	181	181	155	155
   q7	2137	2009	1808	1808
   q8	2649	2876	2751	2751
   q9	7608	7528	7486	7486
   q10	3122	3305	2827	2827
   q11	629	530	491	491
   q12	699	806	638	638
   q13	3644	4004	3508	3508
   q14	277	304	281	281
   q15	549	501	535	501
   q16	1095	1147	1037	1037
   q17	1259	1807	1284	1284
   q18	7878	7777	7687	7687
   q19	811	1194	1056	1056
   q20	1960	1973	1829	1829
   q21	4806	4461	4357	4357
   q22	1122	1060	1002	1002
   Total cold run time: 53987 ms
   Total hot run time: 52762 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3579491236

   # Cloud UT Coverage Report
   Increment line coverage ` ` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/2462a11bc42619b4fdcffd67b314243593f9dc64_2462a11bc42619b4fdcffd67b314243593f9dc64_cloud/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/2462a11bc42619b4fdcffd67b314243593f9dc64_2462a11bc42619b4fdcffd67b314243593f9dc64_cloud/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 80.86% (1707/2111) |
   | Line Coverage     | 66.88% (29899/44708) |
   | Region Coverage   | 67.42% (14910/22114) |
   | Branch Coverage   | 57.58% (7928/13768) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3579446287

   # FE Regression Coverage Report
   Increment line coverage `  66.67% (4/6)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_98f7f81cbd7268940b4d1428e678607919e50b8c_merge_fe/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_98f7f81cbd7268940b4d1428e678607919e50b8c_merge_fe/report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584273806

   
   <details>
   <summary>ClickBench: <b>Total hot run time: 27.16 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 b87eaf23bd13dcfd21756e0823e75065d36d475b, data reload: false
   
   query1	0.05	0.05	0.04
   query2	0.12	0.05	0.04
   query3	0.26	0.08	0.09
   query4	1.60	0.12	0.11
   query5	0.28	0.25	0.25
   query6	1.16	0.65	0.64
   query7	0.03	0.03	0.03
   query8	0.06	0.04	0.04
   query9	0.59	0.50	0.51
   query10	0.56	0.58	0.56
   query11	0.17	0.11	0.10
   query12	0.16	0.12	0.11
   query13	0.62	0.60	0.61
   query14	0.98	1.01	1.00
   query15	0.82	0.80	0.81
   query16	0.41	0.41	0.42
   query17	1.07	1.04	1.00
   query18	0.23	0.22	0.22
   query19	1.94	1.73	1.73
   query20	0.02	0.02	0.02
   query21	15.45	0.26	0.14
   query22	4.94	0.05	0.05
   query23	15.99	0.27	0.10
   query24	1.21	0.66	0.18
   query25	0.08	0.11	0.05
   query26	0.14	0.13	0.13
   query27	0.05	0.04	0.06
   query28	4.07	1.21	1.02
   query29	12.56	3.90	3.21
   query30	0.29	0.13	0.12
   query31	2.82	0.62	0.40
   query32	3.24	0.54	0.46
   query33	3.07	3.06	3.10
   query34	16.86	5.20	4.52
   query35	4.58	4.56	4.60
   query36	0.67	0.50	0.49
   query37	0.10	0.07	0.06
   query38	0.07	0.04	0.04
   query39	0.05	0.03	0.03
   query40	0.17	0.13	0.13
   query41	0.08	0.04	0.03
   query42	0.04	0.03	0.03
   query43	0.05	0.03	0.04
   Total cold run time: 97.71 s
   Total hot run time: 27.16 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3541658527

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3541792570

   # Cloud UT Coverage Report
   Increment line coverage ` ` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/22249e2f89b2266965806f7444efdeb3e16ab171_22249e2f89b2266965806f7444efdeb3e16ab171_cloud/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/22249e2f89b2266965806f7444efdeb3e16ab171_22249e2f89b2266965806f7444efdeb3e16ab171_cloud/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 80.56% (1683/2089) |
   | Line Coverage     | 66.68% (29684/44518) |
   | Region Coverage   | 67.22% (14808/22028) |
   | Branch Coverage   | 57.52% (7884/13706) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3542033407

   # FE UT Coverage Report
   Increment line coverage `  100.00% (2/2)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_22249e2f89b2266965806f7444efdeb3e16ab171/fe_increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_22249e2f89b2266965806f7444efdeb3e16ab171/fe_report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3540526231

   # FE Regression Coverage Report
   Increment line coverage `  100.00% (2/2)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_2e0bd88b8407116b6be9c54135c09920f79b3756_merge_fe/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_2e0bd88b8407116b6be9c54135c09920f79b3756_merge_fe/report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3544940463

   # FE Regression Coverage Report
   Increment line coverage `  100.00% (2/2)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_22249e2f89b2266965806f7444efdeb3e16ab171_merge_fe/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_22249e2f89b2266965806f7444efdeb3e16ab171_merge_fe/report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3570396205

   
   <details>
   <summary>TPC-DS: <b>Total hot run time: 187790 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 6d4a5a626312a4f4c9a77a0ad5c412dca03620b0, data reload: false
   
   query1	1001	423	387	387
   query2	6555	1679	1666	1666
   query3	6752	215	227	215
   query4	25758	23675	23446	23446
   query5	4929	653	487	487
   query6	346	235	220	220
   query7	4671	495	317	317
   query8	295	248	249	248
   query9	8719	2608	2640	2608
   query10	513	347	287	287
   query11	15625	15152	15184	15152
   query12	186	122	115	115
   query13	1690	570	464	464
   query14	11146	9054	9009	9009
   query15	203	198	172	172
   query16	7597	721	504	504
   query17	1180	749	618	618
   query18	2020	426	319	319
   query19	204	193	168	168
   query20	135	127	123	123
   query21	220	131	113	113
   query22	4087	4322	3986	3986
   query23	33936	32990	32812	32812
   query24	8474	2389	2364	2364
   query25	634	523	460	460
   query26	1236	275	166	166
   query27	2734	497	357	357
   query28	4348	2187	2174	2174
   query29	760	668	483	483
   query30	301	221	196	196
   query31	918	803	724	724
   query32	82	76	73	73
   query33	580	358	324	324
   query34	804	856	524	524
   query35	802	808	716	716
   query36	910	948	897	897
   query37	114	114	93	93
   query38	3478	3521	3440	3440
   query39	1451	1417	1394	1394
   query40	225	131	128	128
   query41	71	62	65	62
   query42	126	112	107	107
   query43	466	485	450	450
   query44	1242	773	772	772
   query45	186	180	171	171
   query46	869	981	648	648
   query47	1778	1817	1724	1724
   query48	419	410	323	323
   query49	777	482	398	398
   query50	652	678	410	410
   query51	3973	3911	3859	3859
   query52	114	113	108	108
   query53	244	259	198	198
   query54	302	291	273	273
   query55	89	91	83	83
   query56	321	325	313	313
   query57	1167	1187	1110	1110
   query58	292	273	275	273
   query59	2474	2551	2506	2506
   query60	342	346	348	346
   query61	160	164	167	164
   query62	795	730	656	656
   query63	233	195	193	193
   query64	4414	1185	937	937
   query65	4030	3935	3924	3924
   query66	1139	454	352	352
   query67	15114	15013	14881	14881
   query68	4607	963	650	650
   query69	503	341	304	304
   query70	1271	1195	1265	1195
   query71	419	328	312	312
   query72	6042	5027	4974	4974
   query73	662	565	354	354
   query74	9295	9071	9043	9043
   query75	3279	3303	2812	2812
   query76	3307	1129	707	707
   query77	499	409	329	329
   query78	9557	9730	8828	8828
   query79	1894	865	603	603
   query80	723	583	525	525
   query81	518	258	225	225
   query82	240	166	130	130
   query83	278	275	250	250
   query84	265	105	89	89
   query85	896	488	446	446
   query86	384	308	292	292
   query87	3669	3711	3572	3572
   query88	2783	2274	2243	2243
   query89	383	338	294	294
   query90	1826	223	221	221
   query91	195	163	147	147
   query92	79	68	69	68
   query93	1236	1009	690	690
   query94	638	446	339	339
   query95	418	321	323	321
   query96	491	550	290	290
   query97	2858	2950	2871	2871
   query98	246	214	204	204
   query99	1284	1405	1272	1272
   Total cold run time: 268153 ms
   Total hot run time: 187790 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3579366961

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3579917789

   
   <details>
   <summary>TPC-DS: <b>Total hot run time: 185569 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 2462a11bc42619b4fdcffd67b314243593f9dc64, data reload: false
   
   query1	1040	411	396	396
   query2	6585	1694	1699	1694
   query3	6762	241	231	231
   query4	25815	22632	22811	22632
   query5	5789	655	480	480
   query6	347	279	250	250
   query7	4654	492	306	306
   query8	308	276	264	264
   query9	8741	2574	2561	2561
   query10	576	383	327	327
   query11	15298	14789	14605	14605
   query12	187	123	117	117
   query13	1725	616	476	476
   query14	11923	8983	8920	8920
   query15	265	203	187	187
   query16	7853	695	541	541
   query17	1638	777	624	624
   query18	2083	420	337	337
   query19	309	228	192	192
   query20	135	127	123	123
   query21	218	138	114	114
   query22	3927	4051	4044	4044
   query23	33152	32373	32134	32134
   query24	8293	2379	2376	2376
   query25	574	523	443	443
   query26	1218	280	168	168
   query27	2686	490	364	364
   query28	4341	2144	2119	2119
   query29	789	649	521	521
   query30	306	247	221	221
   query31	813	733	638	638
   query32	85	71	69	69
   query33	617	384	339	339
   query34	756	878	533	533
   query35	795	842	762	762
   query36	947	986	869	869
   query37	130	109	88	88
   query38	3354	3387	3305	3305
   query39	1459	1404	1422	1404
   query40	225	127	121	121
   query41	67	61	60	60
   query42	133	125	121	121
   query43	437	447	410	410
   query44	1278	758	760	758
   query45	200	195	183	183
   query46	860	985	626	626
   query47	1732	1749	1678	1678
   query48	438	425	342	342
   query49	808	500	421	421
   query50	665	674	416	416
   query51	3870	3889	3866	3866
   query52	114	112	118	112
   query53	240	264	202	202
   query54	328	303	287	287
   query55	96	101	105	101
   query56	364	352	343	343
   query57	1160	1186	1109	1109
   query58	301	274	280	274
   query59	2763	2908	2768	2768
   query60	351	354	336	336
   query61	165	165	162	162
   query62	766	738	651	651
   query63	236	194	203	194
   query64	4449	1188	912	912
   query65	4068	3950	4030	3950
   query66	1049	454	331	331
   query67	15188	14969	14743	14743
   query68	8272	958	649	649
   query69	547	355	306	306
   query70	1219	1190	1182	1182
   query71	487	349	327	327
   query72	5768	5219	5531	5219
   query73	642	770	371	371
   query74	8899	8490	8569	8490
   query75	3582	3347	2852	2852
   query76	3447	1126	692	692
   query77	750	410	310	310
   query78	9697	9720	8985	8985
   query79	2581	852	592	592
   query80	734	597	502	502
   query81	496	273	240	240
   query82	475	172	133	133
   query83	305	262	256	256
   query84	304	129	100	100
   query85	907	496	441	441
   query86	341	278	273	273
   query87	3558	3466	3400	3400
   query88	3679	2265	2277	2265
   query89	412	339	314	314
   query90	1977	221	209	209
   query91	175	172	142	142
   query92	84	71	65	65
   query93	1637	950	667	667
   query94	752	445	329	329
   query95	477	412	397	397
   query96	507	558	289	289
   query97	2918	2952	2872	2872
   query98	242	216	214	214
   query99	1427	1387	1235	1235
   Total cold run time: 276142 ms
   Total hot run time: 185569 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3579944771

   
   <details>
   <summary>ClickBench: <b>Total hot run time: 28.41 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 2462a11bc42619b4fdcffd67b314243593f9dc64, data reload: false
   
   query1	0.07	0.06	0.05
   query2	0.15	0.07	0.07
   query3	0.33	0.10	0.09
   query4	1.61	0.11	0.11
   query5	0.28	0.26	0.27
   query6	1.19	0.66	0.67
   query7	0.03	0.03	0.03
   query8	0.07	0.06	0.06
   query9	0.60	0.54	0.51
   query10	0.59	0.58	0.58
   query11	0.24	0.12	0.12
   query12	0.24	0.13	0.13
   query13	0.64	0.61	0.60
   query14	1.00	0.98	1.00
   query15	0.92	0.82	0.82
   query16	0.41	0.40	0.40
   query17	1.10	1.09	1.03
   query18	0.10	0.09	0.08
   query19	2.01	1.90	1.92
   query20	0.01	0.01	0.02
   query21	15.44	0.27	0.23
   query22	5.53	0.09	0.10
   query23	15.80	0.36	0.21
   query24	1.07	0.39	0.27
   query25	0.10	0.09	0.09
   query26	0.19	0.18	0.18
   query27	0.09	0.10	0.10
   query28	3.37	1.29	1.10
   query29	12.56	3.89	3.19
   query30	0.33	0.13	0.11
   query31	2.80	0.61	0.43
   query32	3.24	0.57	0.48
   query33	3.05	3.15	3.31
   query34	15.80	5.25	4.64
   query35	4.62	4.66	4.69
   query36	0.60	0.49	0.49
   query37	0.23	0.08	0.08
   query38	0.18	0.06	0.06
   query39	0.06	0.05	0.05
   query40	0.17	0.16	0.15
   query41	0.10	0.06	0.06
   query42	0.09	0.05	0.05
   query43	0.07	0.05	0.05
   Total cold run time: 97.08 s
   Total hot run time: 28.41 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3580550803

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584006257

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584184783

   # FE UT Coverage Report
   Increment line coverage `  80.00% (8/10)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_b87eaf23bd13dcfd21756e0823e75065d36d475b/fe_increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_b87eaf23bd13dcfd21756e0823e75065d36d475b/fe_report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3575703703

   
   <details>
   <summary>TPC-DS: <b>Total hot run time: 185792 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 6f1aaee068839444ae7dd524bf84bcce15c4fba9, data reload: false
   
   query1	1047	411	422	411
   query2	6583	1718	1731	1718
   query3	6759	232	224	224
   query4	26269	22556	22644	22556
   query5	4435	650	476	476
   query6	344	268	265	265
   query7	4641	486	302	302
   query8	300	253	248	248
   query9	8692	2574	2617	2574
   query10	537	362	316	316
   query11	15105	14970	14651	14651
   query12	177	115	110	110
   query13	1684	605	496	496
   query14	10792	9074	9077	9074
   query15	214	205	193	193
   query16	7843	709	525	525
   query17	1665	788	603	603
   query18	2050	428	316	316
   query19	252	217	189	189
   query20	126	118	120	118
   query21	223	135	115	115
   query22	3927	4078	3869	3869
   query23	33235	32124	32391	32124
   query24	8464	2461	2399	2399
   query25	559	513	448	448
   query26	1226	276	163	163
   query27	2709	488	374	374
   query28	4342	2140	2132	2132
   query29	775	651	505	505
   query30	309	250	214	214
   query31	840	727	634	634
   query32	81	74	65	65
   query33	590	375	331	331
   query34	750	884	527	527
   query35	783	850	750	750
   query36	934	1000	866	866
   query37	125	108	89	89
   query38	3315	3351	3317	3317
   query39	1457	1418	1434	1418
   query40	223	126	116	116
   query41	66	61	61	61
   query42	125	114	111	111
   query43	451	494	444	444
   query44	1229	762	753	753
   query45	202	190	190	190
   query46	889	991	636	636
   query47	1736	1782	1682	1682
   query48	421	432	353	353
   query49	772	496	414	414
   query50	659	673	412	412
   query51	3980	4031	3899	3899
   query52	106	110	104	104
   query53	250	263	193	193
   query54	322	297	275	275
   query55	87	81	87	81
   query56	343	346	315	315
   query57	1166	1176	1097	1097
   query58	297	295	275	275
   query59	2735	2984	2766	2766
   query60	362	349	343	343
   query61	165	187	165	165
   query62	751	720	665	665
   query63	235	198	194	194
   query64	4417	1209	883	883
   query65	4074	4019	3994	3994
   query66	1092	428	335	335
   query67	15299	14914	14866	14866
   query68	5178	922	656	656
   query69	524	351	294	294
   query70	1274	1244	1283	1244
   query71	423	337	315	315
   query72	5816	5311	5528	5311
   query73	642	730	387	387
   query74	8606	8681	8410	8410
   query75	3364	3344	2841	2841
   query76	3313	1150	727	727
   query77	517	393	335	335
   query78	9666	9770	8961	8961
   query79	2794	846	629	629
   query80	820	606	581	581
   query81	532	278	231	231
   query82	706	168	129	129
   query83	303	264	249	249
   query84	310	113	105	105
   query85	920	489	440	440
   query86	397	282	285	282
   query87	3527	3509	3421	3421
   query88	3786	2216	2218	2216
   query89	386	334	302	302
   query90	1861	223	216	216
   query91	174	170	145	145
   query92	81	71	61	61
   query93	2574	938	659	659
   query94	763	424	328	328
   query95	479	397	404	397
   query96	512	589	309	309
   query97	2905	2960	2839	2839
   query98	226	213	203	203
   query99	1441	1380	1242	1242
   Total cold run time: 271433 ms
   Total hot run time: 185792 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3542373223

   
   <details>
   <summary>ClickBench: <b>Total hot run time: 28.32 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 22249e2f89b2266965806f7444efdeb3e16ab171, data reload: false
   
   query1	0.05	0.04	0.05
   query2	0.13	0.07	0.06
   query3	0.32	0.07	0.08
   query4	1.60	0.08	0.09
   query5	0.26	0.26	0.27
   query6	1.18	0.64	0.64
   query7	0.03	0.03	0.03
   query8	0.08	0.06	0.07
   query9	0.58	0.51	0.51
   query10	0.58	0.55	0.56
   query11	0.24	0.13	0.14
   query12	0.26	0.15	0.15
   query13	0.63	0.62	0.62
   query14	1.00	0.98	0.99
   query15	0.85	0.81	0.80
   query16	0.40	0.38	0.39
   query17	1.04	1.04	1.10
   query18	0.23	0.22	0.22
   query19	1.87	1.81	1.79
   query20	0.02	0.02	0.01
   query21	15.40	0.27	0.24
   query22	4.98	0.10	0.09
   query23	15.38	0.37	0.22
   query24	2.18	0.46	0.35
   query25	0.09	0.09	0.09
   query26	0.18	0.18	0.17
   query27	0.09	0.09	0.08
   query28	4.01	1.26	1.08
   query29	12.54	3.96	3.26
   query30	0.35	0.12	0.12
   query31	2.82	0.63	0.42
   query32	3.23	0.58	0.50
   query33	3.03	3.12	3.08
   query34	16.27	5.18	4.59
   query35	4.58	4.62	4.62
   query36	0.61	0.51	0.48
   query37	0.24	0.09	0.08
   query38	0.19	0.06	0.06
   query39	0.06	0.05	0.04
   query40	0.19	0.18	0.16
   query41	0.12	0.06	0.07
   query42	0.07	0.06	0.06
   query43	0.07	0.06	0.05
   Total cold run time: 98.03 s
   Total hot run time: 28.32 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3575232569

   # Cloud UT Coverage Report
   Increment line coverage ` ` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/6f1aaee068839444ae7dd524bf84bcce15c4fba9_6f1aaee068839444ae7dd524bf84bcce15c4fba9_cloud/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/6f1aaee068839444ae7dd524bf84bcce15c4fba9_6f1aaee068839444ae7dd524bf84bcce15c4fba9_cloud/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 80.86% (1707/2111) |
   | Line Coverage     | 66.86% (29891/44708) |
   | Region Coverage   | 67.41% (14906/22114) |
   | Branch Coverage   | 57.57% (7926/13768) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3575649266

   
   <details>
   <summary>TPC-H: <b>Total hot run time: 36199 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 6f1aaee068839444ae7dd524bf84bcce15c4fba9, data reload: false
   
   ------ Round 1 ----------------------------------
   q1	17467	5495	5139	5139
   q2	2024	329	211	211
   q3	10187	1225	729	729
   q4	10235	904	377	377
   q5	7540	2470	2156	2156
   q6	194	170	136	136
   q7	956	783	645	645
   q8	9349	1434	1263	1263
   q9	7513	5414	5402	5402
   q10	6941	2288	1835	1835
   q11	501	294	294	294
   q12	336	365	225	225
   q13	17793	3650	3061	3061
   q14	233	235	209	209
   q15	589	516	507	507
   q16	1035	1011	961	961
   q17	611	885	364	364
   q18	8116	7647	7691	7647
   q19	1116	988	591	591
   q20	387	367	246	246
   q21	4364	3852	3170	3170
   q22	1180	1104	1031	1031
   Total cold run time: 108667 ms
   Total hot run time: 36199 ms
   
   ----- Round 2, with runtime_filter_mode=off -----
   q1	5399	5487	5373	5373
   q2	329	411	340	340
   q3	2514	2824	2492	2492
   q4	1455	1887	1441	1441
   q5	4585	5087	5005	5005
   q6	173	168	126	126
   q7	2060	2013	1790	1790
   q8	2849	2839	2705	2705
   q9	7753	7795	7444	7444
   q10	3046	3201	2685	2685
   q11	578	505	476	476
   q12	621	730	581	581
   q13	3278	3674	3095	3095
   q14	271	284	259	259
   q15	537	502	502	502
   q16	1045	1079	1027	1027
   q17	1204	1714	1230	1230
   q18	7373	7270	7111	7111
   q19	798	1152	1022	1022
   q20	1947	2003	1836	1836
   q21	4859	4550	4391	4391
   q22	1139	1073	1028	1028
   Total cold run time: 53813 ms
   Total hot run time: 51959 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3540083234

   
   <details>
   <summary>TPC-H: <b>Total hot run time: 34675 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 2e0bd88b8407116b6be9c54135c09920f79b3756, data reload: false
   
   ------ Round 1 ----------------------------------
   q1	17471	5677	5249	5249
   q2	2077	313	207	207
   q3	10157	1275	714	714
   q4	10223	914	383	383
   q5	7503	2445	2098	2098
   q6	192	178	139	139
   q7	936	768	630	630
   q8	9350	1386	1278	1278
   q9	7419	5404	5391	5391
   q10	6925	2272	1782	1782
   q11	503	300	285	285
   q12	339	378	233	233
   q13	17800	3725	3028	3028
   q14	233	237	222	222
   q15	564	517	520	517
   q16	1027	1005	929	929
   q17	625	885	354	354
   q18	7744	7111	7167	7111
   q19	1099	971	583	583
   q20	370	361	247	247
   q21	3896	2490	2305	2305
   q22	1103	1039	990	990
   Total cold run time: 107556 ms
   Total hot run time: 34675 ms
   
   ----- Round 2, with runtime_filter_mode=off -----
   q1	5417	5344	5321	5321
   q2	277	319	236	236
   q3	2247	2724	2352	2352
   q4	1392	1806	1380	1380
   q5	4364	5138	5032	5032
   q6	188	170	135	135
   q7	2012	1986	1869	1869
   q8	2662	2890	2773	2773
   q9	7550	7544	7466	7466
   q10	3039	3304	2811	2811
   q11	593	531	522	522
   q12	682	793	633	633
   q13	3549	3996	3347	3347
   q14	307	295	302	295
   q15	533	501	517	501
   q16	1064	1234	1088	1088
   q17	1278	1802	1329	1329
   q18	7941	7842	7666	7666
   q19	837	844	1196	844
   q20	1924	1985	1822	1822
   q21	4817	4472	4407	4407
   q22	1094	1087	1036	1036
   Total cold run time: 53767 ms
   Total hot run time: 52865 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3539858549

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3544904228

   # BE Regression && UT Coverage Report
   Increment line coverage `97.22% (35/36)` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_22249e2f89b2266965806f7444efdeb3e16ab171_merge/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_22249e2f89b2266965806f7444efdeb3e16ab171_merge/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 71.51% (24347/34046) |
   | Line Coverage     | 57.98% (253031/436409) |
   | Region Coverage   | 53.17% (210489/395881) |
   | Branch Coverage   | 54.57% (89951/164834) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2562538354


##########
be/src/pipeline/exec/scan_operator.cpp:
##########
@@ -72,6 +72,38 @@ bool ScanLocalState<Derived>::should_run_serial() const {
     return _parent->cast<typename Derived::Parent>()._should_run_serial;
 }
 
+int ScanLocalStateBase::max_scanners_concurrency(RuntimeState* state) const {
+    // For select * from table limit 10; should just use one thread.
+    if (should_run_serial()) {
+        return 1;
+    }
+    /*
+     * The max concurrency of scanners for each ScanLocalStateBase is determined by:
+     * 1. User specified max_scanners_concurrency which is set through session variable.
+     * 2. Default: std::max(48, CpuInfo::num_cores() * 2))
+     *
+     * If this is a serial operator, the max concurrency should multiply by the number of parallel instances of the operator.
+     */
+    return (state->max_scanners_concurrency() ? state->max_scanners_concurrency()
+                                              : std::max(48, CpuInfo::num_cores() * 2) * 2) *
+           (state->query_parallel_instance_num() / _parent->parallelism(state));
+}
+
+int ScanLocalStateBase::min_scanners_concurrency(RuntimeState* state) const {
+    /*
+     * The min concurrency of scanners for each ScanLocalStateBase is determined by:
+     * 1. User specified min_scan_concurrency_of_scan_scheduler which is set through session variable.
+     * 2. Default: 2 * state->get_query_ctx()->get_scan_scheduler()->get_max_threads()
+     */
+    return state->min_scan_concurrency_of_scan_scheduler()
+                   ? state->min_scan_concurrency_of_scan_scheduler()
+                   : 2 * state->get_query_ctx()->get_scan_scheduler()->get_max_threads();
+}
+
+vectorized::ScannerScheduler* ScanLocalStateBase::scan_scheduler(RuntimeState* state) const {
+    return state->get_query_ctx()->get_scan_scheduler();

Review Comment:
   这个默认使用remote scan scheduler; 然后再olap scan operator 里,返回local scan scheduler



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584006210

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "Gabriel39 (via GitHub)" <gi...@apache.org>.
Gabriel39 commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584006687

   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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584236606

   
   <details>
   <summary>TPC-H: <b>Total hot run time: 34513 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 b87eaf23bd13dcfd21756e0823e75065d36d475b, data reload: false
   
   ------ Round 1 ----------------------------------
   q1	17405	5135	4954	4954
   q2	2060	320	214	214
   q3	10147	1299	735	735
   q4	10259	881	371	371
   q5	7525	2519	2246	2246
   q6	180	167	137	137
   q7	949	813	635	635
   q8	9347	1365	1032	1032
   q9	7147	5505	5371	5371
   q10	6882	2223	1828	1828
   q11	529	322	296	296
   q12	358	375	232	232
   q13	17769	3668	3044	3044
   q14	235	226	210	210
   q15	591	521	509	509
   q16	1050	1023	954	954
   q17	606	773	506	506
   q18	8026	7101	7060	7060
   q19	1090	961	590	590
   q20	343	342	231	231
   q21	3564	3696	2396	2396
   q22	1003	968	962	962
   Total cold run time: 107065 ms
   Total hot run time: 34513 ms
   
   ----- Round 2, with runtime_filter_mode=off -----
   q1	5155	4955	4913	4913
   q2	339	397	309	309
   q3	2209	2707	2277	2277
   q4	1316	1756	1360	1360
   q5	4267	4593	4621	4593
   q6	224	175	130	130
   q7	2077	2009	1823	1823
   q8	2706	2516	2473	2473
   q9	7546	7504	7473	7473
   q10	3082	3342	2814	2814
   q11	634	517	493	493
   q12	660	735	595	595
   q13	3593	3936	3294	3294
   q14	297	328	290	290
   q15	556	504	512	504
   q16	1111	1132	1085	1085
   q17	1166	1473	1412	1412
   q18	7848	7782	7643	7643
   q19	803	818	821	818
   q20	2018	2041	1906	1906
   q21	4879	4337	4213	4213
   q22	1037	1013	981	981
   Total cold run time: 53523 ms
   Total hot run time: 51399 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3575724065

   
   <details>
   <summary>ClickBench: <b>Total hot run time: 28.51 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 6f1aaee068839444ae7dd524bf84bcce15c4fba9, data reload: false
   
   query1	0.05	0.04	0.04
   query2	0.12	0.07	0.06
   query3	0.30	0.08	0.07
   query4	1.61	0.09	0.09
   query5	0.28	0.27	0.27
   query6	1.18	0.66	0.65
   query7	0.03	0.02	0.03
   query8	0.08	0.06	0.07
   query9	0.58	0.54	0.52
   query10	0.58	0.58	0.58
   query11	0.23	0.13	0.13
   query12	0.25	0.14	0.15
   query13	0.63	0.61	0.62
   query14	0.99	0.99	0.99
   query15	0.85	0.81	0.80
   query16	0.40	0.38	0.39
   query17	1.05	1.09	1.02
   query18	0.20	0.20	0.19
   query19	1.85	1.85	1.89
   query20	0.02	0.01	0.02
   query21	15.40	0.28	0.24
   query22	4.97	0.10	0.10
   query23	15.40	0.38	0.22
   query24	2.26	0.43	0.33
   query25	0.10	0.08	0.08
   query26	0.19	0.18	0.18
   query27	0.09	0.09	0.09
   query28	3.73	1.28	1.09
   query29	12.54	3.92	3.22
   query30	0.33	0.13	0.11
   query31	2.80	0.60	0.43
   query32	3.24	0.58	0.50
   query33	3.04	3.11	3.11
   query34	16.50	5.22	4.65
   query35	4.63	4.68	4.66
   query36	0.59	0.50	0.50
   query37	0.23	0.09	0.09
   query38	0.21	0.06	0.06
   query39	0.06	0.05	0.05
   query40	0.20	0.17	0.16
   query41	0.11	0.06	0.06
   query42	0.07	0.05	0.05
   query43	0.06	0.06	0.05
   Total cold run time: 98.03 s
   Total hot run time: 28.51 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3581091087

   
   <details>
   <summary>ClickBench: <b>Total hot run time: 28.37 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 dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0, data reload: false
   
   query1	0.05	0.04	0.04
   query2	0.14	0.07	0.06
   query3	0.32	0.07	0.08
   query4	1.60	0.10	0.10
   query5	0.26	0.25	0.24
   query6	1.18	0.65	0.62
   query7	0.03	0.02	0.03
   query8	0.07	0.07	0.06
   query9	0.58	0.52	0.50
   query10	0.55	0.56	0.56
   query11	0.26	0.14	0.14
   query12	0.27	0.14	0.14
   query13	0.65	0.63	0.62
   query14	1.01	1.01	1.00
   query15	0.88	0.81	0.81
   query16	0.38	0.38	0.40
   query17	1.02	1.05	0.99
   query18	0.23	0.22	0.22
   query19	1.86	1.77	1.76
   query20	0.02	0.01	0.02
   query21	15.39	0.28	0.23
   query22	4.98	0.10	0.09
   query23	15.42	0.37	0.22
   query24	2.41	0.43	0.29
   query25	0.10	0.09	0.09
   query26	0.19	0.17	0.17
   query27	0.10	0.09	0.09
   query28	3.78	1.36	1.18
   query29	12.60	3.98	3.32
   query30	0.31	0.13	0.11
   query31	2.80	0.61	0.45
   query32	3.26	0.57	0.49
   query33	3.08	3.09	3.12
   query34	16.77	5.14	4.57
   query35	4.70	4.70	4.67
   query36	0.62	0.49	0.48
   query37	0.24	0.09	0.08
   query38	0.21	0.06	0.05
   query39	0.07	0.05	0.05
   query40	0.20	0.17	0.16
   query41	0.11	0.06	0.06
   query42	0.08	0.05	0.04
   query43	0.06	0.05	0.06
   Total cold run time: 98.84 s
   Total hot run time: 28.37 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2562538984


##########
be/src/pipeline/exec/scan_operator.cpp:
##########
@@ -72,6 +72,38 @@ bool ScanLocalState<Derived>::should_run_serial() const {
     return _parent->cast<typename Derived::Parent>()._should_run_serial;
 }
 
+int ScanLocalStateBase::max_scanners_concurrency(RuntimeState* state) const {
+    // For select * from table limit 10; should just use one thread.
+    if (should_run_serial()) {
+        return 1;
+    }
+    /*
+     * The max concurrency of scanners for each ScanLocalStateBase is determined by:
+     * 1. User specified max_scanners_concurrency which is set through session variable.
+     * 2. Default: std::max(48, CpuInfo::num_cores() * 2))
+     *
+     * If this is a serial operator, the max concurrency should multiply by the number of parallel instances of the operator.
+     */
+    return (state->max_scanners_concurrency() ? state->max_scanners_concurrency()
+                                              : std::max(48, CpuInfo::num_cores() * 2) * 2) *
+           (state->query_parallel_instance_num() / _parent->parallelism(state));
+}
+
+int ScanLocalStateBase::min_scanners_concurrency(RuntimeState* state) const {
+    /*
+     * The min concurrency of scanners for each ScanLocalStateBase is determined by:
+     * 1. User specified min_scan_concurrency_of_scan_scheduler which is set through session variable.
+     * 2. Default: 2 * state->get_query_ctx()->get_scan_scheduler()->get_max_threads()
+     */
+    return state->min_scan_concurrency_of_scan_scheduler()

Review Comment:
   min_scan_concurrency_of_scan_scheduler > 0



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3578950312

   
   <details>
   <summary>ClickBench: <b>Total hot run time: 28.47 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 98f7f81cbd7268940b4d1428e678607919e50b8c, data reload: false
   
   query1	0.05	0.05	0.05
   query2	0.13	0.07	0.06
   query3	0.31	0.08	0.08
   query4	1.60	0.10	0.09
   query5	0.28	0.26	0.26
   query6	1.18	0.66	0.64
   query7	0.03	0.03	0.02
   query8	0.12	0.06	0.06
   query9	0.59	0.54	0.52
   query10	0.57	0.57	0.58
   query11	0.23	0.14	0.12
   query12	0.25	0.14	0.14
   query13	0.64	0.62	0.61
   query14	1.00	0.99	0.98
   query15	0.85	0.81	0.81
   query16	0.40	0.39	0.39
   query17	1.07	1.03	1.02
   query18	0.20	0.16	0.17
   query19	1.91	1.93	1.91
   query20	0.02	0.01	0.01
   query21	15.39	0.28	0.23
   query22	4.97	0.11	0.11
   query23	15.39	0.38	0.22
   query24	2.31	0.45	0.34
   query25	0.09	0.10	0.09
   query26	0.19	0.18	0.18
   query27	0.09	0.09	0.08
   query28	3.92	1.31	1.10
   query29	12.54	3.88	3.25
   query30	0.32	0.13	0.13
   query31	2.80	0.62	0.42
   query32	3.24	0.58	0.50
   query33	3.06	3.09	3.11
   query34	16.68	5.19	4.64
   query35	4.62	4.59	4.60
   query36	0.59	0.51	0.50
   query37	0.23	0.08	0.08
   query38	0.19	0.05	0.06
   query39	0.06	0.05	0.05
   query40	0.20	0.17	0.16
   query41	0.11	0.06	0.06
   query42	0.07	0.05	0.05
   query43	0.06	0.06	0.05
   Total cold run time: 98.55 s
   Total hot run time: 28.47 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2563711411


##########
be/src/pipeline/exec/file_scan_operator.cpp:
##########
@@ -31,6 +31,39 @@
 
 namespace doris::pipeline {
 #include "common/compile_check_begin.h"
+
+int FileScanLocalState::max_scanners_concurrency(RuntimeState* state) const {
+    // For select * from table limit 10; should just use one thread.
+    if (should_run_serial()) {
+        return 1;
+    }
+    /*
+     * The max concurrency of file scanners for each FileScanLocalState is determined by:
+     * 1. User specified max_file_scanners_concurrency which is set through session variable.
+     * 2. Default: std::max(512, CpuInfo::num_cores() * 10)
+     *
+     * If this is a serial operator, the max concurrency should multiply by the number of parallel instances of the operator.
+     */
+    return (state->max_file_scanners_concurrency() ? state->max_file_scanners_concurrency()
+                                                   : std::max(512, CpuInfo::num_cores() * 10) * 2) *
+           (state->query_parallel_instance_num() / _parent->parallelism(state));
+}
+
+int FileScanLocalState::min_scanners_concurrency(RuntimeState* state) const {

Review Comment:
   rename to min_scan_concurrency_of_scan_scheduler



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei commented on code in PR #58073:
URL: https://github.com/apache/doris/pull/58073#discussion_r2563750292


##########
be/src/pipeline/exec/scan_operator.cpp:
##########
@@ -72,6 +72,38 @@ bool ScanLocalState<Derived>::should_run_serial() const {
     return _parent->cast<typename Derived::Parent>()._should_run_serial;
 }
 
+int ScanLocalStateBase::max_scanners_concurrency(RuntimeState* state) const {
+    // For select * from table limit 10; should just use one thread.
+    if (should_run_serial()) {
+        return 1;
+    }
+    /*
+     * The max concurrency of scanners for each ScanLocalStateBase is determined by:
+     * 1. User specified max_scanners_concurrency which is set through session variable.
+     * 2. Default: std::max(48, CpuInfo::num_cores() * 2))
+     *
+     * If this is a serial operator, the max concurrency should multiply by the number of parallel instances of the operator.
+     */
+    return (state->max_scanners_concurrency() ? state->max_scanners_concurrency()
+                                              : std::max(48, CpuInfo::num_cores() * 2) * 2) *
+           (state->query_parallel_instance_num() / _parent->parallelism(state));
+}
+
+int ScanLocalStateBase::min_scanners_concurrency(RuntimeState* state) const {

Review Comment:
   rename to min_scan_concurrency_of_scan_scheduler



-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3580440587

   # BE UT Coverage Report
   Increment line coverage `19.70% (13/66)` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/2462a11bc42619b4fdcffd67b314243593f9dc64_2462a11bc42619b4fdcffd67b314243593f9dc64/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/2462a11bc42619b4fdcffd67b314243593f9dc64_2462a11bc42619b4fdcffd67b314243593f9dc64/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 52.91% (18434/34840) |
   | Line Coverage     | 38.39% (168468/438813) |
   | Region Coverage   | 33.16% (130724/394165) |
   | Branch Coverage   | 34.12% (56282/164948) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3580803158

   # FE UT Coverage Report
   Increment line coverage `  80.00% (8/10)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0/fe_increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0/fe_report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3581024548

   
   <details>
   <summary>TPC-H: <b>Total hot run time: 35169 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 dd41d8a649b2599efab873ad3bd0f9aa54ca9fc0, data reload: false
   
   ------ Round 1 ----------------------------------
   q1	17430	4991	5003	4991
   q2	2028	320	204	204
   q3	10174	1309	746	746
   q4	10239	892	374	374
   q5	7571	2513	2212	2212
   q6	195	173	138	138
   q7	968	790	640	640
   q8	9353	1369	1010	1010
   q9	7382	5276	5380	5276
   q10	6885	2250	1797	1797
   q11	527	304	291	291
   q12	377	369	231	231
   q13	17779	3657	3053	3053
   q14	236	234	218	218
   q15	588	514	514	514
   q16	1066	1012	943	943
   q17	633	834	438	438
   q18	7950	7509	7546	7509
   q19	1249	970	638	638
   q20	387	363	229	229
   q21	4237	4007	2675	2675
   q22	1107	1096	1042	1042
   Total cold run time: 108361 ms
   Total hot run time: 35169 ms
   
   ----- Round 2, with runtime_filter_mode=off -----
   q1	5123	5123	5139	5123
   q2	325	395	337	337
   q3	2314	3033	2458	2458
   q4	1371	1896	1370	1370
   q5	4504	4490	4473	4473
   q6	203	164	125	125
   q7	1960	1905	1791	1791
   q8	2653	2472	2450	2450
   q9	7622	7394	7565	7394
   q10	3212	3238	2764	2764
   q11	567	501	475	475
   q12	642	699	555	555
   q13	3253	3646	3017	3017
   q14	262	284	262	262
   q15	539	496	490	490
   q16	1032	1043	1009	1009
   q17	1098	1357	1390	1357
   q18	7270	7140	6925	6925
   q19	777	768	748	748
   q20	1884	1950	1795	1795
   q21	4637	4229	4045	4045
   q22	1114	1068	966	966
   Total cold run time: 52362 ms
   Total hot run time: 49929 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3575400198

   # FE UT Coverage Report
   Increment line coverage `  66.67% (4/6)` :tada:
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/58073_6f1aaee068839444ae7dd524bf84bcce15c4fba9/fe_increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/58073_6f1aaee068839444ae7dd524bf84bcce15c4fba9/fe_report/index.html)
   


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "hello-stephen (via GitHub)" <gi...@apache.org>.
hello-stephen commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3584108713

   # Cloud UT Coverage Report
   Increment line coverage ` ` :tada:
   
   [Increment coverage report](http://coverage.selectdb-in.cc/coverage/b87eaf23bd13dcfd21756e0823e75065d36d475b_b87eaf23bd13dcfd21756e0823e75065d36d475b_cloud/increment_report/index.html)
   [Complete coverage report](http://coverage.selectdb-in.cc/coverage/b87eaf23bd13dcfd21756e0823e75065d36d475b_b87eaf23bd13dcfd21756e0823e75065d36d475b_cloud/report/index.html)
   | Category          | Coverage           |
   |-------------------|--------------------|
   | Function Coverage | 80.86% (1707/2111) |
   | Line Coverage     | 66.84% (29882/44708) |
   | Region Coverage   | 67.38% (14901/22114) |
   | Branch Coverage   | 57.59% (7929/13768) |


-- 
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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "doris-robot (via GitHub)" <gi...@apache.org>.
doris-robot commented on PR #58073:
URL: https://github.com/apache/doris/pull/58073#issuecomment-3570414432

   
   <details>
   <summary>ClickBench: <b>Total hot run time: 28.34 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 6d4a5a626312a4f4c9a77a0ad5c412dca03620b0, data reload: false
   
   query1	0.05	0.05	0.04
   query2	0.13	0.06	0.06
   query3	0.33	0.08	0.08
   query4	1.61	0.09	0.09
   query5	0.26	0.25	0.24
   query6	1.18	0.65	0.63
   query7	0.03	0.02	0.03
   query8	0.07	0.06	0.06
   query9	0.59	0.53	0.51
   query10	0.57	0.55	0.55
   query11	0.25	0.12	0.14
   query12	0.25	0.14	0.14
   query13	0.64	0.62	0.61
   query14	1.01	0.99	1.01
   query15	0.89	0.83	0.82
   query16	0.39	0.39	0.40
   query17	0.96	1.08	1.05
   query18	0.22	0.20	0.21
   query19	1.86	1.77	1.80
   query20	0.02	0.01	0.01
   query21	15.43	0.27	0.23
   query22	4.98	0.09	0.09
   query23	15.39	0.38	0.22
   query24	2.34	0.45	0.31
   query25	0.09	0.09	0.10
   query26	0.18	0.17	0.18
   query27	0.09	0.08	0.08
   query28	3.55	1.26	1.09
   query29	12.56	3.93	3.32
   query30	0.33	0.13	0.12
   query31	2.82	0.60	0.44
   query32	3.23	0.58	0.50
   query33	3.08	3.13	3.02
   query34	16.38	5.29	4.63
   query35	4.68	4.67	4.69
   query36	0.62	0.52	0.48
   query37	0.23	0.09	0.09
   query38	0.20	0.06	0.06
   query39	0.06	0.05	0.04
   query40	0.20	0.18	0.16
   query41	0.11	0.06	0.06
   query42	0.07	0.06	0.05
   query43	0.06	0.05	0.04
   Total cold run time: 97.99 s
   Total hot run time: 28.34 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] [Improvement](external) Increase scanner concurrency [doris]

Posted by "yiguolei (via GitHub)" <gi...@apache.org>.
yiguolei merged PR #58073:
URL: https://github.com/apache/doris/pull/58073


-- 
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