Skip to content

branch-3.1: [Opt](mow) Opt mow load performance and CPU usage #55073#55767

Merged
morrySnow merged 4 commits intoapache:branch-3.1from
bobhan1:branch-3.1-pick-55073
Sep 13, 2025
Merged

branch-3.1: [Opt](mow) Opt mow load performance and CPU usage #55073#55767
morrySnow merged 4 commits intoapache:branch-3.1from
bobhan1:branch-3.1-pick-55073

Conversation

@bobhan1
Copy link
Contributor

@bobhan1 bobhan1 commented Sep 8, 2025 •
edited
Loading

@bobhan1 bobhan1 requested a review from morrySnow as a code owner September 8, 2025 06:43
@Thearas
Copy link
Contributor

Thearas commented Sep 8, 2025

Thank you for your contribution to Apache Doris.
Don't know what should be done next? See 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?

@bobhan1
Copy link
Contributor Author

bobhan1 commented Sep 8, 2025

run buildall

dataroaring
dataroaring previously approved these changes Sep 8, 2025
View reviewed changes
Copy link
Contributor

@dataroaring dataroaring left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@morrySnow
Copy link
Contributor

morrySnow commented Sep 8, 2025

run buildall

@doris-robot
Copy link

doris-robot commented Sep 9, 2025

TPC-H: Total hot run time: 32899 ms 
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/tpch-tools
Tpch sf100 test result on commit 109edcfbd6779e523b967f696d51b3da9f110b33, data reload: false

------ Round 1 ----------------------------------
q1	17685	5473	5525	5473
q2	2028	399	281	281
q3	12143	1253	756	756
q4	10256	873	449	449
q5	8600	2402	2183	2183
q6	181	164	131	131
q7	902	763	607	607
q8	9340	1457	1263	1263
q9	5243	5016	5012	5012
q10	6763	2300	1802	1802
q11	480	284	251	251
q12	337	365	216	216
q13	17772	3583	3009	3009
q14	223	216	209	209
q15	544	475	467	467
q16	420	436	372	372
q17	614	884	365	365
q18	7100	6487	6327	6327
q19	1531	961	561	561
q20	328	336	197	197
q21	2733	2195	1960	1960
q22	1076	1033	1008	1008
Total cold run time: 106299 ms
Total hot run time: 32899 ms

----- Round 2, with runtime_filter_mode=off -----
q1	5554	5529	5557	5529
q2	247	332	235	235
q3	2251	2657	2307	2307
q4	1327	1756	1333	1333
q5	4435	5049	5004	5004
q6	178	162	128	128
q7	2102	1972	1840	1840
q8	2660	2907	2741	2741
q9	7213	7289	7232	7232
q10	3064	3337	2676	2676
q11	562	490	493	490
q12	630	758	622	622
q13	3424	3785	3123	3123
q14	283	294	264	264
q15	518	474	469	469
q16	461	478	433	433
q17	1244	1756	1266	1266
q18	7559	7520	7265	7265
q19	821	1191	1049	1049
q20	2027	2068	1923	1923
q21	5377	4989	4789	4789
q22	1075	1032	1025	1025
Total cold run time: 53012 ms
Total hot run time: 51743 ms

@doris-robot
Copy link

doris-robot commented Sep 9, 2025

TPC-DS: Total hot run time: 193546 ms 
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/tpcds-tools
TPC-DS sf100 test result on commit 109edcfbd6779e523b967f696d51b3da9f110b33, data reload: false

query1	957	407	406	406
query2	6242	1925	1894	1894
query3	8698	209	206	206
query4	33522	24373	23920	23920
query5	4367	607	496	496
query6	307	192	183	183
query7	4248	496	308	308
query8	299	239	240	239
query9	9180	2542	2546	2542
query10	504	330	257	257
query11	18183	15364	15210	15210
query12	156	108	102	102
query13	1551	547	436	436
query14	8918	7518	7516	7516
query15	243	187	185	185
query16	8137	669	522	522
query17	1585	794	605	605
query18	2199	416	337	337
query19	217	210	180	180
query20	133	129	125	125
query21	213	124	109	109
query22	4539	4574	4442	4442
query23	34900	34451	34314	34314
query24	7346	2697	2703	2697
query25	562	506	423	423
query26	1266	304	175	175
query27	1995	494	365	365
query28	5422	2205	2201	2201
query29	799	622	465	465
query30	241	197	166	166
query31	1023	944	863	863
query32	121	57	60	57
query33	518	375	308	308
query34	757	849	516	516
query35	815	815	727	727
query36	1017	1057	944	944
query37	105	95	71	71
query38	4008	4073	4022	4022
query39	1541	1468	1470	1468
query40	220	120	108	108
query41	52	49	48	48
query42	129	107	106	106
query43	502	501	486	486
query44	1329	818	816	816
query45	184	176	170	170
query46	887	1079	673	673
query47	1983	1957	1880	1880
query48	399	419	345	345
query49	792	483	424	424
query50	682	690	447	447
query51	7371	7514	7401	7401
query52	105	103	94	94
query53	232	268	189	189
query54	565	543	476	476
query55	82	84	79	79
query56	274	291	275	275
query57	1263	1262	1235	1235
query58	234	223	221	221
query59	3044	3209	2929	2929
query60	291	273	266	266
query61	114	119	113	113
query62	805	742	683	683
query63	230	192	192	192
query64	4580	1000	653	653
query65	3408	3282	3310	3282
query66	950	413	311	311
query67	16371	15826	15409	15409
query68	7758	808	530	530
query69	486	304	273	273
query70	1217	1107	1118	1107
query71	373	295	269	269
query72	5787	3752	3755	3752
query73	631	746	346	346
query74	10219	9160	9010	9010
query75	3217	3160	2674	2674
query76	3096	1199	773	773
query77	572	359	282	282
query78	10366	10441	9712	9712
query79	3365	885	594	594
query80	607	530	436	436
query81	500	258	217	217
query82	564	115	89	89
query83	179	167	145	145
query84	277	104	78	78
query85	790	372	299	299
query86	339	320	284	284
query87	4257	4340	4313	4313
query88	4746	2389	2378	2378
query89	414	331	309	309
query90	1792	193	195	193
query91	138	147	108	108
query92	65	56	53	53
query93	1895	881	540	540
query94	677	404	285	285
query95	332	277	264	264
query96	491	612	277	277
query97	3204	3265	3125	3125
query98	216	211	204	204
query99	1504	1423	1314	1314
Total cold run time: 293986 ms
Total hot run time: 193546 ms

@doris-robot
Copy link

doris-robot commented Sep 9, 2025

ClickBench: Total hot run time: 28.88 s 
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/clickbench-tools
ClickBench test result on commit 109edcfbd6779e523b967f696d51b3da9f110b33, data reload: false

query1	0.03	0.03	0.03
query2	0.07	0.03	0.03
query3	0.23	0.07	0.06
query4	1.63	0.10	0.10
query5	0.51	0.52	0.51
query6	1.13	0.72	0.72
query7	0.02	0.02	0.01
query8	0.05	0.03	0.04
query9	0.60	0.50	0.50
query10	0.55	0.57	0.55
query11	0.15	0.11	0.10
query12	0.14	0.11	0.11
query13	0.62	0.59	0.60
query14	0.76	0.80	0.81
query15	0.85	0.84	0.84
query16	0.41	0.38	0.39
query17	1.00	1.12	1.07
query18	0.24	0.23	0.22
query19	1.98	1.85	1.87
query20	0.02	0.01	0.01
query21	15.38	0.97	0.56
query22	0.75	0.76	0.66
query23	15.15	1.42	0.58
query24	2.92	1.49	0.87
query25	0.21	0.09	0.13
query26	0.36	0.16	0.13
query27	0.04	0.04	0.03
query28	12.96	1.01	0.43
query29	12.58	3.97	3.26
query30	0.25	0.09	0.07
query31	2.83	0.60	0.38
query32	3.23	0.55	0.45
query33	3.01	2.99	3.03
query34	16.63	5.23	4.54
query35	4.62	4.54	4.57
query36	0.66	0.49	0.47
query37	0.09	0.05	0.06
query38	0.05	0.04	0.03
query39	0.04	0.02	0.03
query40	0.16	0.14	0.14
query41	0.08	0.03	0.03
query42	0.04	0.03	0.02
query43	0.03	0.03	0.04
Total cold run time: 103.06 s
Total hot run time: 28.88 s

@doris-robot
Copy link

doris-robot commented Sep 9, 2025

BE UT Coverage Report

Increment line coverage 16.76% (29/173) 🎉

Increment coverage report
Complete coverage report

Category Coverage
Function Coverage 45.53% (12758/28023)
Line Coverage 36.37% (113787/312863)
Region Coverage 33.99% (65067/191407)
Branch Coverage 31.03% (34153/110070)

@hello-stephen
Copy link
Contributor

hello-stephen commented Sep 9, 2025

BE Regression && UT Coverage Report

Increment line coverage 90.17% (156/173) 🎉

Increment coverage report
Complete coverage report

Category Coverage
Function Coverage 76.34% (21042/27562)
Line Coverage 69.68% (217313/311889)
Region Coverage 67.67% (130046/192184)
Branch Coverage 61.19% (67671/110584)

@bobhan1
[Opt](mow) Opt mow load performance and CPU usage (apache#55073)
d76ba97 
This PR let mow load use delete bitmap cache if possible when lookup row
key to reduce CPU usage.

Add 2 configs
`enable_prefill_output_dbm_agg_cache_after_compaction`/`enable_prefill_all_dbm_agg_cache_after_compaction`(default
true) to control whether to fresh delete bitmap cache on corresponding
rowsets after compaction.

None

- 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:
apache/doris-website#1214 -->

- [ ] Confirm the release note
- [ ] Confirm test cases
- [ ] Confirm document
- [ ] Add branch pick label <!-- Add branch pick label that this PR
should merge into -->
@bobhan1
[Fix](case) Comment some case to avoid p0 stuck (apache#55733)
2facc89 
fix apache#55073
@bobhan1
fix
86791e0
@bobhan1
tmp
215667a
@bobhan1 bobhan1 force-pushed the branch-3.1-pick-55073 branch from 109edcf to 215667a Compare September 12, 2025 10:42
@bobhan1
Copy link
Contributor Author

bobhan1 commented Sep 12, 2025

run buildall

@doris-robot
Copy link

doris-robot commented Sep 12, 2025

TPC-H: Total hot run time: 32905 ms 
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/tpch-tools
Tpch sf100 test result on commit 215667a11dad9b60d38aff95defe54c192dceb9b, data reload: false

------ Round 1 ----------------------------------
q1	17811	5499	5579	5499
q2	2035	409	279	279
q3	12407	1260	747	747
q4	10316	878	451	451
q5	9359	2381	2186	2186
q6	185	164	133	133
q7	887	737	611	611
q8	9328	1445	1160	1160
q9	5289	4979	4946	4946
q10	6768	2293	1798	1798
q11	480	284	265	265
q12	333	350	204	204
q13	17784	3594	3052	3052
q14	230	238	208	208
q15	539	458	459	458
q16	413	429	368	368
q17	599	883	354	354
q18	6866	6516	6516	6516
q19	1201	964	538	538
q20	329	348	206	206
q21	2892	2162	1941	1941
q22	1036	1010	985	985
Total cold run time: 107087 ms
Total hot run time: 32905 ms

----- Round 2, with runtime_filter_mode=off -----
q1	5587	5461	5524	5461
q2	244	324	229	229
q3	2282	2653	2275	2275
q4	1364	1818	1353	1353
q5	4449	5078	4999	4999
q6	168	167	129	129
q7	2116	1961	1811	1811
q8	2643	2853	2706	2706
q9	7310	7314	7247	7247
q10	3053	3213	2734	2734
q11	565	528	493	493
q12	677	779	630	630
q13	3435	3858	3160	3160
q14	279	300	284	284
q15	512	482	453	453
q16	450	488	445	445
q17	1236	1742	1250	1250
q18	7699	7497	7257	7257
q19	815	1187	1077	1077
q20	2008	2051	1894	1894
q21	5345	4766	4586	4586
q22	1093	1098	968	968
Total cold run time: 53330 ms
Total hot run time: 51441 ms

@doris-robot
Copy link

doris-robot commented Sep 12, 2025

TPC-DS: Total hot run time: 191960 ms 
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/tpcds-tools
TPC-DS sf100 test result on commit 215667a11dad9b60d38aff95defe54c192dceb9b, data reload: false

query1	972	408	400	400
query2	6151	1906	1855	1855
query3	8681	198	194	194
query4	33671	24182	23269	23269
query5	3831	584	448	448
query6	307	204	179	179
query7	4200	496	325	325
query8	301	248	233	233
query9	9352	2577	2589	2577
query10	463	326	258	258
query11	18314	15440	15248	15248
query12	157	106	105	105
query13	1567	548	453	453
query14	9874	6678	7572	6678
query15	261	192	186	186
query16	8108	646	495	495
query17	1570	788	592	592
query18	2180	426	334	334
query19	236	184	173	173
query20	133	126	128	126
query21	214	131	117	117
query22	4487	4545	4302	4302
query23	35223	34379	34145	34145
query24	7791	2699	2705	2699
query25	561	506	436	436
query26	1175	291	178	178
query27	2055	488	366	366
query28	5117	2211	2181	2181
query29	792	595	463	463
query30	240	200	168	168
query31	1001	956	828	828
query32	93	61	59	59
query33	525	380	336	336
query34	763	882	521	521
query35	785	830	740	740
query36	1034	1084	979	979
query37	106	96	63	63
query38	4054	4008	4022	4008
query39	1513	1509	1438	1438
query40	209	118	111	111
query41	51	51	51	51
query42	118	104	111	104
query43	522	517	485	485
query44	1385	826	835	826
query45	191	184	170	170
query46	886	1049	674	674
query47	1994	2021	1957	1957
query48	417	448	354	354
query49	790	487	430	430
query50	678	700	432	432
query51	7425	7397	7255	7255
query52	105	104	95	95
query53	231	261	206	206
query54	545	552	479	479
query55	79	81	85	81
query56	273	260	258	258
query57	1267	1276	1240	1240
query58	224	214	223	214
query59	2961	3154	2926	2926
query60	298	293	265	265
query61	140	140	112	112
query62	815	765	719	719
query63	236	195	196	195
query64	4547	1015	652	652
query65	3360	3348	3329	3329
query66	1046	414	313	313
query67	16376	15711	15425	15425
query68	7405	814	537	537
query69	492	310	272	272
query70	1170	1148	1136	1136
query71	375	288	258	258
query72	5779	3840	3924	3840
query73	632	745	348	348
query74	10404	9132	9296	9132
query75	3156	3153	2672	2672
query76	3087	1155	774	774
query77	475	379	267	267
query78	10385	10535	9663	9663
query79	3018	892	608	608
query80	642	525	434	434
query81	528	258	223	223
query82	552	119	89	89
query83	183	158	143	143
query84	291	104	77	77
query85	787	359	313	313
query86	349	324	284	284
query87	4336	4302	4255	4255
query88	5096	2429	2388	2388
query89	414	332	304	304
query90	1844	189	191	189
query91	136	141	109	109
query92	66	56	53	53
query93	1782	914	543	543
query94	706	424	287	287
query95	347	276	271	271
query96	491	599	279	279
query97	3228	3311	3213	3213
query98	225	208	207	207
query99	1508	1411	1279	1279
Total cold run time: 294916 ms
Total hot run time: 191960 ms

@doris-robot
Copy link

doris-robot commented Sep 12, 2025

ClickBench: Total hot run time: 28.68 s 
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/clickbench-tools
ClickBench test result on commit 215667a11dad9b60d38aff95defe54c192dceb9b, data reload: false

query1	0.03	0.03	0.03
query2	0.07	0.02	0.03
query3	0.23	0.07	0.07
query4	1.63	0.10	0.11
query5	0.53	0.50	0.53
query6	1.16	0.76	0.75
query7	0.03	0.02	0.02
query8	0.04	0.03	0.03
query9	0.58	0.51	0.50
query10	0.59	0.55	0.55
query11	0.14	0.11	0.10
query12	0.13	0.11	0.11
query13	0.61	0.59	0.60
query14	0.78	0.79	0.79
query15	0.83	0.84	0.83
query16	0.37	0.39	0.39
query17	1.01	1.02	0.98
query18	0.23	0.22	0.22
query19	1.95	1.88	1.82
query20	0.02	0.01	0.01
query21	15.41	1.00	0.58
query22	0.74	0.74	0.69
query23	15.16	1.44	0.60
query24	3.16	1.40	0.62
query25	0.11	0.14	0.24
query26	0.36	0.18	0.13
query27	0.06	0.04	0.04
query28	12.54	0.99	0.44
query29	12.57	3.90	3.24
query30	0.24	0.10	0.06
query31	2.83	0.59	0.37
query32	3.22	0.53	0.47
query33	2.98	3.05	3.01
query34	16.47	5.17	4.56
query35	4.58	4.56	4.55
query36	0.65	0.51	0.47
query37	0.08	0.06	0.06
query38	0.04	0.03	0.04
query39	0.03	0.03	0.02
query40	0.16	0.13	0.12
query41	0.09	0.02	0.02
query42	0.03	0.02	0.02
query43	0.04	0.03	0.03
Total cold run time: 102.51 s
Total hot run time: 28.68 s

@doris-robot
Copy link

doris-robot commented Sep 12, 2025

BE UT Coverage Report

Increment line coverage 18.08% (32/177) 🎉

Increment coverage report
Complete coverage report

Category Coverage
Function Coverage 45.52% (12763/28037)
Line Coverage 36.36% (113797/313004)
Region Coverage 33.99% (65084/191478)
Branch Coverage 31.02% (34154/110108)

@hello-stephen
Copy link
Contributor

hello-stephen commented Sep 12, 2025

BE Regression && UT Coverage Report

Increment line coverage 90.40% (160/177) 🎉

Increment coverage report
Complete coverage report

Category Coverage
Function Coverage 76.58% (21116/27575)
Line Coverage 69.90% (218097/312033)
Region Coverage 67.84% (130427/192255)
Branch Coverage 61.40% (67927/110622)

@morrySnow morrySnow merged commit 4c30059 into apache:branch-3.1 Sep 13, 2025
21 of 22 checks passed
@morrySnow morrySnow mentioned this pull request Sep 22, 2025
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@morrySnow morrySnow morrySnow approved these changes

@dataroaring dataroaring dataroaring left review comments

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

6 participants

@bobhan1 @Thearas @morrySnow @doris-robot @hello-stephen @dataroaring

Comments