Commit Graph

123550 Commits

Author SHA1 Message Date
vdimir
b9bb4af6be
Support setting up fail points in config 2023-08-31 11:45:39 +00:00
vdimir
4fa9dcb326
Fix cluster discovery initialization 2023-08-31 11:33:01 +00:00
Jiebin Sun
7c529e5691
Optimize the merge if all hashSets are singleLevel in UniqExactSet (#52973)
* Optimize the merge if all hashSets are singleLevel

In PR(https://github.com/ClickHouse/ClickHouse/pull/50748), it has added new phase
`parallelizeMergePrepare` before merge if all the hashSets are not all singleLevel
or not all twoLevel. Then it will convert all the singleLevelSet to twoLevelSet in
parallel, which will increase the CPU utilization and QPS.

But if all the hashtables are singleLevel, it could also benefit from the
`parallelizeMergePrepare` optimization in most cases if the hashtable size are not
too small. By tuning the Query `SELECT COUNT(DISTINCT SearchPhase) FROM hits_v1`
in different threads, we have got the mild threshold 6,000.

Test patch with the Query 'SELECT COUNT(DISTINCT Title) FROM hits_v1' on 2x80 vCPUs
server. If the threads are less than 48, the hashSets are all twoLevel or mixed by
singleLevel and twoLevel. If the threads are over 56, all the hashSets are singleLevel.
And the QPS has got at most 2.35x performance gain.

Threads	Opt/Base
8	100.0%
16	99.4%
24	110.3%
32	99.9%
40	99.3%
48	99.8%
56	183.0%
64	234.7%
72	233.1%
80	229.9%
88	224.5%
96	229.6%
104	235.1%
112	229.5%
120	229.1%
128	217.8%
136	222.9%
144	217.8%
152	204.3%
160	203.2%

Signed-off-by: Jiebin Sun <jiebin.sun@intel.com>

* Add the comment and explanation for PR#52973

Signed-off-by: Jiebin Sun <jiebin.sun@intel.com>

---------

Signed-off-by: Jiebin Sun <jiebin.sun@intel.com>
2023-08-30 11:26:16 +02:00
Antonio Andelic
36fb7cfbd1
Merge pull request #54012 from ClickHouse/refactor-async-insert-with-dedup
Refactor logic around async insert with deduplication
2023-08-30 09:37:55 +02:00
Nikita Mikhaylov
97efd14e90
Merge pull request #54009 from ClickHouse/change-istorage-interface
Change IStorage interface by random walk, no goal in particular
2023-08-29 22:22:53 +02:00
Alexey Milovidov
9b74939110
Merge pull request #53742 from aiven-sal/aiven-sal/fix_53187
transform: correctly handle default column with multiple rows
2023-08-29 21:57:59 +03:00
Alexey Milovidov
62747ea20f
Merge pull request #53739 from Hexta/arm-clocksource
Do not warn about arch_sys_counter clock
2023-08-29 21:57:39 +03:00
Sergei Trifonov
802579f3f1
Merge pull request #49618 from ClickHouse/concurrency-control-controllable
Make concurrency control controllable
2023-08-29 19:44:51 +02:00
Alexey Milovidov
36a01253a0
Merge pull request #54044 from ClickHouse/alexey-milovidov-patch-7
More assertive
2023-08-29 20:33:38 +03:00
Alexey Milovidov
c19a20cc23
More assertive 2023-08-29 20:33:03 +03:00
Antonio Andelic
e96c665be7 Refactor logic around async insert with dedup 2023-08-29 14:48:43 +00:00
Alexander Tokmakov
17f3754193
Merge pull request #53851 from ClickHouse/distributed_no_flush_on_shutdown
Follow-up to #53501
2023-08-29 16:34:22 +02:00
Nikita Mikhaylov
96cabcae3a Change interface 2023-08-29 16:26:48 +02:00
Robert Schulze
342c78bfe6
Merge pull request #53763 from rschu1ze/branch-references
Remove branch references from .gitmodules
2023-08-29 15:29:43 +02:00
Alexander Tokmakov
464c49fdcb
Merge pull request #53929 from ClickHouse/revert-53713-ADQM-1107
Revert "Fixed wrong python test name pattern"
2023-08-29 15:23:01 +02:00
Alexander Tokmakov
e610a1867c
Revert "Fixed wrong python test name pattern" 2023-08-29 15:19:13 +02:00
robot-clickhouse-ci-1
f5c0890b32
Merge pull request #53895 from ClickHouse/vdimir/write_buffer_to_file_segment_dtor
Finalize file descriptor in ~WriteBufferToFileSegment
2023-08-29 15:15:44 +02:00
Anton Popov
6ce4a82a04
Merge pull request #53677 from VanDarkholme7/mutations_subcolumns
Fix bug on mutations with subcolumns of type JSON in predicates of UPDATE and DELETE queries.
2023-08-29 15:08:00 +02:00
Kseniia Sumarokova
463a197da4
Merge pull request #53847 from kssenii/fix-segfault-in-pg
Fix possible segfault while using PostgreSQL engine
2023-08-29 14:45:46 +02:00
Alexey Milovidov
3dc4194978
Merge pull request #53906 from ClickHouse/tzfix
Fix flakiness of 00514_interval_operators
2023-08-29 15:02:41 +03:00
Robert Schulze
93eec131ae
Merge pull request #53920 from rschu1ze/docs-laion
Dataset docs: Update + fix LAION-400M tutorial
2023-08-29 12:59:16 +02:00
robot-ch-test-poll1
2c8cc7979f
Merge pull request #53571 from ClickHouse/password-complexity-req
Add password complexity requirements to docs
2023-08-29 12:42:30 +02:00
Mikhail f. Shiryaev
25eda7f358
Merge pull request #53713 from arenadata/ADQM-1107
Fixed wrong python test name pattern
2023-08-29 12:38:12 +02:00
Robert Schulze
43367f99fb
Fix style 2023-08-29 12:35:56 +02:00
vdimir
a6abe23173
Merge pull request #53913 from cfanbo/patch-1 2023-08-29 12:26:32 +02:00
Robert Schulze
b4219886b4
Dataset docs: Update + fix LAION-400M tutorial 2023-08-29 10:17:13 +00:00
vdimir
9f990965e7
Merge pull request #53909 from ClickHouse/docs-dynamic-column-selection 2023-08-29 12:05:31 +02:00
Chen768959
a2d451d6e6
domainRFC support ipv6(square brackets) and fix #53179 (#53506) 2023-08-29 11:42:59 +02:00
Robert Schulze
679a983094
Switch paragraphs 2023-08-29 10:57:35 +02:00
János Benjamin Antal
a1a45ee905
Merge pull request #53766 from ilejn/exceptions_kafka_consumers
system.kafka_consumers exception info improvements
2023-08-29 10:22:26 +02:00
robot-ch-test-poll
5ccc75b0f7
Merge pull request #53769 from ClickHouse/pufit/keeper-client-node-from-config
KeeperClient auto-detection of keeper node from clickhouse configs
2023-08-29 09:33:11 +02:00
cfanbo
2e49255f06
docs: fix func name error 2023-08-29 15:11:28 +08:00
VanDarkholme7
714bfdd5dc restart CI 2023-08-29 03:16:50 +00:00
Justin de Guzman
be65292714
Rename COLUMNS expression to dynamic column selection 2023-08-28 19:55:42 -07:00
pufit
7f4fc3744d resolve issues 2023-08-28 22:23:01 -04:00
Alexey Milovidov
cb831a5fb7
Merge pull request #53876 from davvard/master
Fix: USearch deserialize
2023-08-29 04:45:30 +03:00
Justin de Guzman
bc6c8ea410
Add info about password complexity req for ClickHouse Cloud 2023-08-28 18:16:53 -07:00
robot-ch-test-poll4
ea7f583708
Merge pull request #53891 from ClibMouse/feature/uint128-hash-big-endian
Make UInt128TrivialHash endianness-independent
2023-08-29 02:55:47 +02:00
Alexey Milovidov
0c0e8e548a
Merge pull request #53864 from jsc0218/ignore-foreign-keys-in-tables-definition
Ignore foreign keys in tables definition
2023-08-29 03:09:54 +03:00
Alexey Milovidov
9b4ded6804
Merge pull request #53756 from ClibMouse/update_openSSL_3.0.10
Upgrade openSSL to  v3.0.10
2023-08-29 03:00:36 +03:00
Michael Kolupaev
ad71bf5f7d Fix flakiness of 00514_interval_operators 2023-08-28 23:35:47 +00:00
Alexey Milovidov
e09abfc011
Merge pull request #52600 from xiaolei565/xuelei_dev
The configuration of S3 endpoint allow using it from the root, and append '/' automatically if needed.
2023-08-29 01:42:08 +03:00
Alexey Gerasimchuck
1069244d17 Fix in ci-runner.py 2023-08-28 22:05:36 +00:00
Alexey Milovidov
bbef3ceeb0
Merge pull request #53902 from HarryLeeIBM/hlee-s390x-stripe-log
Fix StripeLog storage endian issue for s390x
2023-08-29 00:35:48 +03:00
Robert Schulze
9f9c4293fa
Update usearch to v2.0.2 2023-08-28 20:10:49 +00:00
HarryLeeIBM
dcecf52a68 Fix StripeLog storage endian issue for s390x 2023-08-28 11:35:04 -07:00
vdimir
9f158d4d63
Finalize file descriptor in ~WriteBufferToFileSegment 2023-08-28 16:01:10 +00:00
Alexander Tokmakov
98e6a6b09b
Merge pull request #53849 from ClickHouse/tavplubix-patch-9
Fix segfault in `TableNameHints` (with `Lazy` database)
2023-08-28 17:51:30 +02:00
ltrk2
e52934ce61 Make UInt128TrivialHash endianness-independent 2023-08-28 08:39:50 -07:00
Alexander Tokmakov
d41eca1dcc rename new method 2023-08-28 16:01:00 +02:00