Skip to content

[fix](catalog) set timeout for split fetch#39346

Merged
morningman merged 2 commits intoapache:masterfrom
morningman:split_cache
Aug 16, 2024
Merged

[fix](catalog) set timeout for split fetch#39346
morningman merged 2 commits intoapache:masterfrom
morningman:split_cache

Conversation

@morningman
Copy link
Copy Markdown
Contributor

@morningman morningman commented Aug 14, 2024

When fetch splits in batch, BE will send rpc to FE to fetch batch of splits.
The FE may be blocked when listing file from hdfs, causing BE block too.
This PR add timeout on FE side to avoid BE block.

@doris-robot
Copy link
Copy Markdown

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

Since 2024-03-18, the Document has been moved to doris-website.
See Doris Document.

@morningman
Copy link
Copy Markdown
Contributor Author

run buildall

@doris-robot
Copy link
Copy Markdown

TPC-H: Total hot run time: 38144 ms
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/tpch-tools
Tpch sf100 test result on commit adaec717144643d992047a956647d93eaa693439, data reload: false

------ Round 1 ----------------------------------
q1	18036	4549	4405	4405
q2	2984	181	178	178
q3	11224	1215	1143	1143
q4	10318	915	752	752
q5	7802	2894	2854	2854
q6	231	137	137	137
q7	975	605	624	605
q8	9542	2075	2064	2064
q9	7330	6532	6582	6532
q10	7040	2291	2197	2197
q11	527	247	255	247
q12	393	227	218	218
q13	18892	2996	2998	2996
q14	289	237	243	237
q15	525	489	487	487
q16	511	388	382	382
q17	983	674	730	674
q18	7650	6781	6820	6781
q19	4696	1051	1101	1051
q20	709	321	330	321
q21	3768	2882	3046	2882
q22	1099	1025	1001	1001
Total cold run time: 115524 ms
Total hot run time: 38144 ms

----- Round 2, with runtime_filter_mode=off -----
q1	4534	4275	4298	4275
q2	380	278	276	276
q3	2867	2657	2649	2649
q4	1939	1706	1612	1612
q5	5402	5419	5412	5412
q6	224	136	133	133
q7	2057	1679	1708	1679
q8	3203	3355	3359	3355
q9	8499	8455	8466	8455
q10	3413	3157	3142	3142
q11	597	503	524	503
q12	801	619	590	590
q13	17419	2979	2988	2979
q14	310	275	270	270
q15	526	478	493	478
q16	478	416	424	416
q17	1801	1513	1505	1505
q18	7811	7520	7374	7374
q19	1667	1649	1529	1529
q20	2015	1809	1835	1809
q21	5112	4980	5056	4980
q22	1100	1036	1012	1012
Total cold run time: 72155 ms
Total hot run time: 54433 ms

@doris-robot
Copy link
Copy Markdown

TPC-DS: Total hot run time: 184102 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 adaec717144643d992047a956647d93eaa693439, data reload: false

query1	914	388	370	370
query2	6460	1931	1849	1849
query3	6650	207	217	207
query4	33780	23158	23156	23156
query5	4248	509	514	509
query6	282	182	175	175
query7	4598	295	304	295
query8	250	205	199	199
query9	8640	2440	2418	2418
query10	430	283	265	265
query11	17657	15193	15041	15041
query12	145	95	97	95
query13	1616	365	357	357
query14	9670	6243	6951	6243
query15	239	165	167	165
query16	7638	476	479	476
query17	1540	557	550	550
query18	1902	278	269	269
query19	188	139	139	139
query20	110	103	104	103
query21	206	100	100	100
query22	4339	4166	3895	3895
query23	34509	33462	33262	33262
query24	11698	2825	2812	2812
query25	571	373	371	371
query26	1084	155	150	150
query27	2940	269	273	269
query28	7868	2060	2034	2034
query29	736	400	397	397
query30	305	154	148	148
query31	982	739	768	739
query32	99	56	53	53
query33	744	277	284	277
query34	974	473	470	470
query35	857	729	735	729
query36	1077	935	930	930
query37	140	82	84	82
query38	4008	3896	3793	3793
query39	1437	1381	1378	1378
query40	273	120	112	112
query41	44	47	43	43
query42	117	94	98	94
query43	502	472	462	462
query44	1266	732	738	732
query45	192	163	163	163
query46	1110	773	760	760
query47	1864	1738	1768	1738
query48	363	290	296	290
query49	1121	419	407	407
query50	812	403	403	403
query51	6808	6673	6688	6673
query52	101	94	92	92
query53	253	184	184	184
query54	848	445	443	443
query55	75	72	78	72
query56	264	245	247	245
query57	1150	1062	1075	1062
query58	240	212	214	212
query59	3022	2767	2770	2767
query60	296	268	277	268
query61	94	93	92	92
query62	838	625	667	625
query63	211	182	180	180
query64	5302	2298	1743	1743
query65	3183	3186	3131	3131
query66	960	361	336	336
query67	15233	15003	14773	14773
query68	4538	549	557	549
query69	409	273	262	262
query70	1213	1103	1109	1103
query71	387	279	272	272
query72	6550	2252	1990	1990
query73	777	321	322	321
query74	9175	8826	8723	8723
query75	3415	2694	2763	2694
query76	2791	1017	1038	1017
query77	513	308	308	308
query78	9896	9098	8913	8913
query79	2361	524	514	514
query80	2138	480	481	480
query81	584	218	226	218
query82	839	135	137	135
query83	270	149	157	149
query84	281	84	80	80
query85	2065	276	269	269
query86	506	298	306	298
query87	4502	4242	4149	4149
query88	4041	2309	2318	2309
query89	392	297	287	287
query90	1834	209	196	196
query91	123	97	100	97
query92	62	48	49	48
query93	2413	531	536	531
query94	1027	292	294	292
query95	345	257	253	253
query96	595	271	270	270
query97	3255	3048	3055	3048
query98	226	200	199	199
query99	1800	1266	1256	1256
Total cold run time: 297396 ms
Total hot run time: 184102 ms

@doris-robot
Copy link
Copy Markdown

ClickBench: Total hot run time: 31.11 s
machine: 'aliyun_ecs.c7a.8xlarge_32C64G'
scripts: https://github.com/apache/doris/tree/master/tools/clickbench-tools
ClickBench test result on commit adaec717144643d992047a956647d93eaa693439, data reload: false

query1	0.04	0.04	0.04
query2	0.09	0.04	0.04
query3	0.22	0.05	0.06
query4	1.67	0.09	0.08
query5	0.50	0.49	0.49
query6	1.12	0.73	0.74
query7	0.02	0.01	0.02
query8	0.05	0.04	0.04
query9	0.55	0.47	0.49
query10	0.54	0.54	0.55
query11	0.15	0.11	0.11
query12	0.16	0.12	0.13
query13	0.59	0.60	0.59
query14	0.77	0.77	0.79
query15	0.89	0.81	0.82
query16	0.38	0.35	0.38
query17	0.99	0.98	1.03
query18	0.23	0.21	0.22
query19	1.95	1.73	1.81
query20	0.01	0.01	0.01
query21	15.39	0.76	0.67
query22	4.57	7.03	2.27
query23	18.29	1.41	1.27
query24	2.15	0.23	0.22
query25	0.13	0.08	0.08
query26	0.30	0.23	0.21
query27	0.46	0.23	0.23
query28	13.28	1.03	1.00
query29	12.65	3.32	3.27
query30	0.23	0.06	0.05
query31	2.88	0.39	0.40
query32	3.28	0.48	0.48
query33	2.90	2.97	2.98
query34	17.12	4.48	4.36
query35	4.45	4.47	4.40
query36	0.65	0.48	0.48
query37	0.18	0.16	0.15
query38	0.15	0.16	0.15
query39	0.04	0.04	0.04
query40	0.15	0.11	0.13
query41	0.09	0.05	0.06
query42	0.06	0.05	0.05
query43	0.04	0.04	0.04
Total cold run time: 110.36 s
Total hot run time: 31.11 s

Copy link
Copy Markdown
Contributor

@kaka11chen kaka11chen left a comment

Choose a reason for hiding this comment

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

LGTM

@github-actions
Copy link
Copy Markdown
Contributor

PR approved by at least one committer and no changes requested.

@github-actions github-actions bot added the approved Indicates a PR has been approved by one committer. label Aug 14, 2024
@github-actions
Copy link
Copy Markdown
Contributor

PR approved by anyone and no changes requested.

@morningman morningman merged commit aa30b5a into apache:master Aug 16, 2024
dataroaring pushed a commit that referenced this pull request Aug 17, 2024
When fetch splits in batch, BE will send rpc to FE to fetch batch of
splits.
The FE may be blocked when listing file from hdfs, causing BE block too.
This PR add timeout on FE side to avoid BE block.
morningman added a commit to morningman/doris that referenced this pull request Aug 20, 2024
When fetch splits in batch, BE will send rpc to FE to fetch batch of
splits.
The FE may be blocked when listing file from hdfs, causing BE block too.
This PR add timeout on FE side to avoid BE block.
@gavinchou gavinchou mentioned this pull request Oct 13, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants