thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-12-12 09:22:05 +00:00
Code Issues Packages Projects Releases Wiki Activity
113,400 Commits 9,297 Branches 1,782 Tags 2.1 GiB
6a0e5fad09
Commit Graph

534 Commits

Author SHA1 Message Date
Smita Kulkarni
b70878aa0e Merge branch 'master' into 42192_Lower_parallel_parsing_threads_with_globs 2023-04-24 19:39:36 +02:00
vdimir
b1f0e0d62f Lock clusters_mutex for cluster_discovery 2023-04-19 16:44:54 +02:00
vdimir
d6416a1321 Store clusters from ClusterDiscovery in separate map 2023-04-19 16:44:54 +02:00
Alexander Gololobov
02876ed87d Share PreparedSet-s between task of the same mutation to reduce mem usage 2023-04-14 16:12:18 +02:00
Dmitry Novik
5cc9b46f78 Merge remote-tracking branch 'origin/master' into optimize-compilation 2023-04-13 16:04:09 +02:00
Azat Khuzhin
011480924a Use forward declaration of ThreadPool 
Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
 2023-04-07 11:25:35 +02:00
Vladimir C
e7a5c96353
Merge pull request #48219 from ClickHouse/vdimir/lock_order_47072 2023-04-06 11:54:33 +02:00
robot-clickhouse
3d2c279859
Merge pull request #48242 from azat/better-throttling 
Per-query/per-server throttling for remote IO/local IO/BACKUPs
 2023-04-05 19:32:37 +02:00
vdimir
01ce3baf0f
Rename Context::getTemporaryVolume to getGlobalTemporaryVolume 2023-04-05 13:42:13 +00:00
Nikita Mikhaylov
7b526d3df9
Always communicate with local replica through loopback interface for parallel replicas (#48246) 2023-04-05 13:32:46 +02:00
vdimir
bae1286b63
Update temp_data_on_disk in shared context 2023-04-05 10:07:44 +00:00
Azat Khuzhin
c332d290d8 Keep only one throttler for BACKUPs IO (instead of separate read/write) 
There is no need in separate read/write throttling, because you cannot
write faster then read anyway, and plus this makes the code less cleaner

(and also it will allow avoid implementing throttling backups to S3,
since it does not use common S3 writer).

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
 2023-04-05 09:40:17 +02:00
Azat Khuzhin
218b1f9c29 Add ability to throttle BACKUPs on per-server/backup basis 
Server settings:
- backup_read_bandwidth_for_server
- backup_write_bandwidth_for_server

Query settings:
- backup_read_bandwidth
- backup_write_bandwidth

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
 2023-04-05 09:39:48 +02:00
Azat Khuzhin
a25dd1d348 Add ability to throttle local IO on per-query/server basis 
Server settings:
- max_local_read_bandwidth_for_server
- max_local_write_bandwidth_for_server

Query settings:
- max_local_read_bandwidth
- max_local_write_bandwidth

This is the preparation for adding ability to throttle BACKUPs

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
 2023-04-05 09:39:15 +02:00
Azat Khuzhin
b3406beeb7 Add per-query network throttling 
Controlled with:
- max_remote_read_network_bandwidth
- max_remote_write_network_bandwidth

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
 2023-04-05 09:39:15 +02:00
Azat Khuzhin
abdb682048 Move some methods for Context into module part 
This will be required later, for per-query bandwidth limiting.

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
 2023-04-05 09:39:15 +02:00
Azat Khuzhin
40cb87c2ac Make remote bandwidth settings server settings 
Right now those two settings are applied only at start:
- max_remote_read_network_bandwidth_for_server
- max_remote_write_network_bandwidth_for_server

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
 2023-04-05 09:39:15 +02:00
SmitaRKulkarni
d9c67a3380
Merge branch 'master' into 42192_Lower_parallel_parsing_threads_with_globs 2023-03-30 09:35:03 +02:00
Dmitry Novik
3488a4d84b Improve file includes 2023-03-24 03:44:52 +01:00
Robert Schulze
657aa65446
Cleanup: Untangle CacheBase's constructors 
This prepares for the custom TTLCachePolicy which users should not be
able to configure explicitly.
 2023-03-13 15:11:12 +00:00
Antonio Andelic
737cf8e149 Better 2023-03-03 15:14:49 +00:00
Smita Kulkarni
3e9ab1276a Updated to reduce parallel parsing based on number of files - When reading from multiple files reduce parallel parsing 2023-02-22 14:56:44 +01:00
Smita Kulkarni
3a6dea5e16 When reading from multiple files displace parallel parsing 
Implementation:
* Added a new parameter to getInput & getInputFormat to disable parallel parsing.
* Currently this is used only by StorageFile as we have not seen degradation for other storages reading from multiple paths.
 2023-02-21 17:03:00 +01:00
Alexey Milovidov
993a39e2a1
Merge pull request #46360 from genzgd/http_client_version 
Allow to provide the protocol version by HTTP clients to tune the compatibility of the Native output format.
 2023-02-16 22:17:04 +03:00
Alexey Milovidov
d102c4eafe
Update Context.h 2023-02-14 11:12:40 +03:00
Geoff Genz
be8bf3a6a3
Merge branch 'master' into http_client_version 2023-02-13 08:43:59 -07:00
Sergei Trifonov
297587531b
Merge branch 'master' into round-robin-merge-scheduler 2023-02-11 17:19:27 +01:00
serxa
c58b165b0f add config option to select scheduling policy 2023-02-11 16:18:42 +00:00
kssenii
3067c1d723 Merge remote-tracking branch 'upstream/master' into resubmit-prefetches 2023-02-11 11:36:23 +01:00
Sergei Trifonov
fbedf4d520
Merge branch 'master' into round-robin-merge-scheduler 2023-02-11 10:30:26 +01:00
Geoff Genz
99c3ff53c5 Merge remote-tracking branch 'origin/master' into http_client_version 
# Conflicts:
#	src/Interpreters/Context.cpp
#	src/Interpreters/Context.h
 2023-02-10 04:35:53 -07:00
Geoff Genz
7ed8ed0284 Add support for client_protocol_version sent with HTTP 2023-02-10 03:47:06 -07:00
serxa
a4006ec5a1 add explanation 2023-02-09 19:49:11 +00:00
serxa
cc3c76944b remove redundant code 2023-02-09 19:31:54 +00:00
serxa
f61f95fcec Replace background executor scheduler merges and mutations with round-robin 2023-02-09 18:54:15 +00:00
Vitaly Baranov
39bacfb9dc
Merge branch 'master' into allow-single-disk-instead-of-storage-policy 2023-02-09 13:16:55 +01:00
kssenii
b0b865c32e Resubmit prefetches 2023-02-08 21:26:24 +01:00
Robert Schulze
9b841d67e8
Merge remote-tracking branch 'origin/master' into qc-max_cache_size 2023-02-05 16:47:02 +00:00
Nikita Mikhaylov
33877b5e00
Parallel replicas. Part [2] (#43772) 2023-02-03 14:34:18 +01:00
kssenii
a391de18a1 Merge remote-tracking branch 'upstream/master' into allow-single-disk-instead-of-storage-policy 2023-02-02 20:49:41 +01:00
Robert Schulze
b1889b0284
Make the naming a little less confusing 2023-02-02 19:18:39 +00:00
Robert Schulze
216636f984
Make query cache global settings re-configurable 2023-02-01 10:35:19 +00:00
Robert Schulze
325c6bdf3d
Renaming: "Query Result Cache" --> "Query Cache" 
Reasons:

- The cache will at some point store intermediate results as opposed to
  only query results. We should change the terminology now without
  having to worry about backward compat.

- Equivalent caches in MySQL (1) and Starrocks (2) are called "query
  cache".

- The new name is ca. 13.8% more catchy.

(1) https://dev.mysql.com/doc/refman/5.6/en/query-cache.html
(2) https://docs.starrocks.io/en-us/2.5/using_starrocks/query_cache
 2023-01-31 09:54:34 +00:00
Robert Schulze
099e30ef2a
Merge remote-tracking branch 'origin/master' into query-result-cache 2023-01-16 08:04:49 +00:00
Nikolai Kochetov
4673b3fe1d
Revert "Revert "Custom reading for mutation"" 2023-01-10 16:31:01 +01:00
Alexander Tokmakov
c8ec130be4
Revert "Custom reading for mutation" 2023-01-10 17:51:30 +03:00
Nikolai Kochetov
3e430b1b6e Add comments. Remove unused flag from Context. 2023-01-09 19:52:11 +00:00
kssenii
8fde606768 Resolve some review comments 2023-01-06 16:10:00 +01:00
kssenii
2a030c1dc0 Merge remote-tracking branch 'upstream/master' into allow-single-disk-instead-of-storage-policy 2023-01-06 15:58:27 +01:00
Robert Schulze
e9e04166d9
Merge remote-tracking branch 'origin/master' into query-result-cache 2023-01-02 08:20:27 +00:00