thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-16 04:32:33 +00:00
Code Issues Packages Projects Releases Wiki Activity
56,530 Commits 9,138 Branches 1,774 Tags 2.1 GiB
b2d7790794
Commit Graph

440 Commits

Author SHA1 Message Date
Alexander Kuzmenkov
15a0f14445
Merge pull request #15419 from myrrc/improvement/diff-types-in-avg-weighted 
Allow different types in avgWeighted. Allow avg and avgWeighed to operate on extended integral types.
 2020-11-26 17:16:48 +03:00
Nikolai Kochetov
729272391f
Merge branch 'master' into ip-dict-no-trie 2020-11-25 23:07:19 +03:00
Azat Khuzhin
688cb6b4d9 Update date_time_short perf test for toUnixTimestamp(Date()) 2020-11-25 21:17:11 +03:00
myrrc
420f2489a7 fixed decimal scales calc, updated the tests 2020-11-24 17:07:59 +03:00
myrrc
fbb0e6e6aa Merge remote-tracking branch 'upstream/master' into improvement/diff-types-in-avg-weighted 2020-11-24 16:04:17 +03:00
vdimir
52bc290616
Regenerate ya.make, add format null to ip_trie.xml 2020-11-24 11:20:11 +03:00
Azat Khuzhin
8931d3eb6f Do not use SET via <full_query> in perf tests 
Since if the connection will be closed (by some reason), then the
setting will not be applied after transparent reconnect (since only
native clickhouse-client can do this, since it parses the query, but
perf tests uses python driver).

Just use inplace SETTINGS clause or <settings>.
 2020-11-21 14:02:21 +03:00
Azat Khuzhin
a3116d5614 Tune aggregating_merge_tree_simple_aggregate_function_string to make it faster 2020-11-21 12:08:59 +03:00
Azat Khuzhin
35231662b3 Improve performance of AggregatingMergeTree w/ SimpleAggregateFunction(String) 
While reading from AggregatingMergeTree with
SimpleAggregateFunction(String) in primary key and
optimize_aggregation_in_order perf top shows:

    Samples: 1M of event 'cycles', 4000 Hz, Event count (approx.): 287759760270 lost: 0/0 drop: 0/0
      Children      Self  Shared Object         Symbol
    +   12.64%    11.39%  clickhouse            [.] memcpy
    +    9.08%     0.23%  [unknown]             [.] 0000000000000000
    +    8.45%     8.40%  clickhouse            [.] ProfileEvents::increment    # <-- this, and in debug it has not 0.08x overhead, but 5.8x overhead
    +    7.68%     7.67%  clickhouse            [.] LZ4_compress_fast_extState
    +    5.29%     5.22%  clickhouse            [.] DB::IAggregateFunctionHelper<DB::AggregateFunctionNullUnary<true, true> >::addFree

The reason is obvious, ProfileEvents is atomic counters (and also they
are nested):

<details>

```
    Samples: 7M of event 'cycles', 4000 Hz, Event count (approx.): 450726149337
    ProfileEvents::increment  /usr/bin/clickhouse [Percent: local period]
    Percent│
           │
           │
           │    Disassembly of section .text:
           │
           │    00000000078d8900 <ProfileEvents::increment(unsigned long, unsigned long)@@Base>:
           │    ProfileEvents::increment(unsigned long, unsigned long):
      0.17 │      push  %rbp
      0.00 │      mov   %rsi,%rbp
      0.04 │      push  %rbx
      0.20 │      mov   %rdi,%rbx
      0.17 │      sub   $0x8,%rsp
      0.26 │    → callq DB::CurrentThread::getProfileEvents
           │    ProfileEvents::Counters::increment(unsigned long, unsigned long):
      0.00 │      lea   0x0(,%rbx,8),%rdi
      0.05 │      nop
           │    unsigned long std::__1::__cxx_atomic_fetch_add<unsigned long, unsigned long>(std::__1::__cxx_atomic_base_impl<unsigned long>*, unsigned long, std::__1::memory_order):
      1.02 │      mov   (%rax),%rdx
     97.04 │      lock  add   %rbp,(%rdx,%rdi,1)
           │    ProfileEvents::Counters::increment(unsigned long, unsigned long):
      0.21 │      mov   0x10(%rax),%rax
      0.04 │      test  %rax,%rax
      0.00 │    → jne   78d8920 <ProfileEvents::increment(unsigned long, unsigned long)@@Base+0x20>
           │    ProfileEvents::increment(unsigned long, unsigned long):
      0.38 │      add   $0x8,%rsp
      0.00 │      pop   %rbx
      0.04 │      pop   %rbp
      0.38 │    ← retq
```

</details>

These ProfileEvents was ArenaAllocChunks (it shows ~1.5M events per
second), and the reason is that the table has
SimpleAggregateFunction(String) in PK, which requires Arena.
But most of the time there Arena wasn't even used, so avoid this cost by
re-creating Arena only if it was "used" (i.e. has new chunks).

Another possibility is to avoid populating Arena::head in ctor, but this
will make the Arena code more complex, so for now this was preferred.

Also as a long-term solution it worth looking at implementing them via
RCU (to move the extra overhead out from the write code path into read
side).
 2020-11-19 23:06:12 +03:00
vdimir
36544a45b7
Merge remote-tracking branch 'upstream/master' into ip-dict-no-trie 2020-11-19 18:56:24 +03:00
alesapin
cdceafdd89 Trying to make read_in_order_many_parts more stable 2020-11-19 13:25:39 +03:00
myrrc
dbc5284b73 replaced Memory by MergeTree in the test to get perftests 2020-11-18 12:51:02 +03:00
vdimir
6dcb38db3f
Minor changes in IP dictionary 2020-11-16 21:08:31 +03:00
Alexander Kuzmenkov
b0e660651a cleanup 2020-11-16 19:09:58 +03:00
vdimir
5e0e22301b
Slightly improve ip_dict loading performance, handle v4 to v6 masks in prepossessing, add more tests 2020-11-15 18:55:34 +03:00
vdimir
ca8e4b1607
Minor fixes of ip dict 2020-11-14 20:03:13 +03:00
myrrc
b1c0089b78 Merge remote-tracking branch 'upstream/master' into improvement/diff-types-in-avg-weighted 2020-11-14 16:50:04 +03:00
Alexander Kuzmenkov
e4dd68039e fix 2020-11-13 08:00:43 +03:00
myrrc
a8d8dd7d3e Merge remote-tracking branch 'upstream/master' into improvement/diff-types-in-avg-weighted 2020-11-12 14:50:59 +03:00
Alexander Kuzmenkov
0d10a4990e fix 2020-11-12 11:27:33 +03:00
Alexander Kuzmenkov
91f88c79f9 Merge remote-tracking branch 'origin/master' into HEAD 2020-11-12 11:20:09 +03:00
Alexey Milovidov
f1d721fa74 Marked some perf test queries as short 2020-11-11 19:58:31 +03:00
Alexander Kuzmenkov
f31d7869ac Merge remote-tracking branch 'origin/master' into HEAD 2020-11-11 14:51:35 +03:00
Alexander Kuzmenkov
906b338a1d fixes 2020-11-11 14:51:21 +03:00
Danila Kutenin
c36e6fe378 Fix tests finally 2020-11-10 18:22:59 +03:00
Danila Kutenin
c0308a5d85 Fix tests, I cannot count number of zeros sometimes 2020-11-10 18:21:39 +03:00
Danila Kutenin
847fa9ca3c Merge 2020-11-10 18:15:26 +03:00
Danila Kutenin
8504efde77 Merge 2020-11-10 18:12:49 +03:00
Danila Kutenin
2057198779 Fix performance tests that became very fast 2020-11-10 17:58:05 +03:00
myrrc
e460248624 Merge remote-tracking branch 'upstream/master' into improvement/diff-types-in-avg-weighted 2020-11-10 14:54:24 +03:00
Alexander Kuzmenkov
84fb76bad4 fixes 2020-11-10 07:19:43 +03:00
Alexander Kuzmenkov
030419db90 Try 12 threads max in perf test. 
This will allow us to use more machines, not only Gold 6320.
 2020-11-09 19:14:05 +03:00
alexey-milovidov
e47ed55d32
Merge pull request #14977 from achimbab/indexOf_bloomfilter 
Fix indexOf() to use BloomFilter
 2020-11-09 17:58:42 +03:00
alexey-milovidov
2994499680
Update bloom_filter_select.xml 2020-11-09 17:58:27 +03:00
Alexey Milovidov
ff906369ea Adjust time for perf test 2020-11-08 23:22:50 +03:00
vdimir
81671ceb39
Add tests/performance/ip_trie.xml [wip] 2020-11-08 21:32:22 +03:00
Alexey Milovidov
f07988eec7 Remove perf test of sumbur hash, because we do not care 2020-11-08 21:17:30 +03:00
Alexey Milovidov
7257a5c3d2 Fix perf tests 2020-11-08 16:55:36 +03:00
Alexey Milovidov
f4178dd91f Fix perf tests 2020-11-08 16:55:26 +03:00
alexey-milovidov
c7aa0c851a
Update bloom_filter.xml 2020-11-08 16:51:45 +03:00
alexey-milovidov
d96914b044
Update bloom_filter.xml 2020-11-08 16:51:03 +03:00
Alexey Milovidov
cc012e2cc7 Merge branch 'master' into achimbab-indexOf_bloomfilter 2020-11-08 14:42:03 +03:00
Alexey Milovidov
990f8a23ab Translate comments to english 2020-11-05 21:51:36 +03:00
alexey-milovidov
ee3e289ec2
Merge pull request #15983 from ClickHouse/no_background_pool_no_more 
Replace BackgroundProcessingPool with SchedulePool task and ThreadPool.
 2020-11-05 21:38:22 +03:00
myrrc
ddd40fc4ce fixed the perf test 2020-11-05 12:20:05 +03:00
myrrc
9788065786 Merge remote-tracking branch 'upstream/master' into improvement/diff-types-in-avg-weighted 2020-11-05 08:44:37 +03:00
alexey-milovidov
dbdff0d779
Merge pull request #16661 from ClickHouse/optimize-or-fill 
Optimize -OrNull and -OrDefault aggregate functions by providing specializations
 2020-11-04 23:37:11 +03:00
alexey-milovidov
ca0037d18b
Update group_by_sundy_li.xml 2020-11-04 18:07:18 +03:00
myrrc
43b2d20314 updated the decimal template magic 2020-11-04 16:14:07 +03:00
myrrc
3d70ab7f3b Merge remote-tracking branch 'upstream/master' into improvement/diff-types-in-avg-weighted 2020-11-04 14:29:59 +03:00
Alexey Milovidov
e0d962a3cf Add a test for #8772 2020-11-04 14:22:02 +03:00
Alexey Milovidov
938ef62532 Add a test 2020-11-04 14:00:43 +03:00
Alexey Milovidov
fdd7f31cf7 Add a test 2020-11-03 23:28:23 +03:00
myrrc
5c6bd218ad adjusted the perftest 2020-11-03 18:01:29 +03:00
myrrc
fcbc0fb91e added specialized performance test 2020-11-03 16:12:27 +03:00
Anton Popov
a3a8e18637
Merge branch 'master' into select_final 2020-11-03 00:00:43 +03:00
alesapin
101aa9e931 Merge branch 'master' into no_background_pool_no_more 2020-11-02 10:49:19 +03:00
Alexander Kuzmenkov
e15a8d5ac0 fixup 2020-10-30 21:03:07 +03:00
Alexander Kuzmenkov
7e76b7b510 more fixes 2020-10-30 18:54:38 +03:00
Alexander Kuzmenkov
38726847fc make the test longer 2020-10-30 18:16:30 +03:00
Alexander Kuzmenkov
3529100da9 Adjust ignored perf test changes after NUMA binding 2020-10-30 18:12:15 +03:00
Anton Popov
48185d437a
Update optimized_select_final.xml 2020-10-30 17:48:59 +03:00
Anton Popov
9f0f981642
Update optimized_select_final.xml 2020-10-29 14:03:50 +03:00
Pavel Kruglov
4592c5e59b Add OPTIMIZE in perf test 2020-10-27 14:42:26 +03:00
alesapin
d5d0819698 Merge branch 'master' into no_background_pool_no_more 2020-10-26 12:58:51 +03:00
alesapin
84dbbdb312 Rename test 2020-10-23 13:42:15 +03:00
alesapin
166b3de466 Make insert test faster 2020-10-23 13:31:33 +03:00
Pavel Kruglov
0274c9e9eb Add perf test 2020-10-23 12:05:57 +03:00
alesapin
7a63b373f9 Merge branch 'blind_fix_of_read_in_order' into no_background_pool_no_more 2020-10-22 15:47:11 +03:00
Alexander Kuzmenkov
6293d1dbbe Merge remote-tracking branch 'origin/master' into HEAD 2020-10-21 18:39:59 +03:00
Alexander Kuzmenkov
242ec7e56c make some tests faster 2020-10-20 15:12:06 +03:00
Vasily Nemkov
1cf9781d4e AES: Attempt to make performance tests faster and more stable. 2020-10-20 08:05:39 +03:00
Vasily Nemkov
4124538b56 Merge remote-tracking branch 'upstream/master' into AES_encrypt_decrypt 2020-10-18 00:00:13 +03:00
alesapin
60d14352f8 Merge branch 'master' into no_background_pool_no_more 2020-10-15 16:34:14 +03:00
alexey-milovidov
1d309c34c8
Merge pull request #15497 from filipecaixeta/master 
Add an SQL function formatReadableTimeDelta to format time delta
 2020-10-14 22:36:57 +03:00
Alexander Kuzmenkov
cb15e72229 make some perf tests slower 2020-10-14 18:39:27 +03:00
Vasily Nemkov
a6f36fc53b Performance tests for encrypt and decrypt functions 2020-10-14 16:33:07 +03:00
alesapin
163d33fd21 Add perftest for merges 
(cherry picked from commit e655f009d2)
 2020-10-13 15:12:28 +03:00
Alexey Milovidov
905bb859a5 Add performance test 2020-10-11 18:25:29 +03:00
Artem Zuikov
4fd1db73a1
Faster 256-bit multiplication (#15418) 2020-09-29 20:52:34 +03:00
Alexander Kuzmenkov
8cbfd69adf
Update decimal_casts.xml 2020-09-24 11:35:35 +03:00
Alexander Kuzmenkov
9b29e31d64 Merge remote-tracking branch 'origin/master' into HEAD 2020-09-23 23:58:43 +03:00
alexey-milovidov
3bc09bb867
Merge pull request #15178 from amosbird/dtcf 
Fix performance issue when doing datetime comparison
 2020-09-23 20:33:23 +03:00
Alexander Kuzmenkov
9f137e77e7
Update decimal_casts.xml 2020-09-23 18:41:52 +03:00
Alexander Kuzmenkov
d2427a121f
Update decimal_casts.xml 2020-09-23 10:15:51 +03:00
Amos Bird
ee45395942
Fix datetime comparison 2020-09-23 10:29:55 +08:00
Artem Zuikov
f80ffc4dce restore decimal_casts perf test 2020-09-22 18:56:24 +03:00
Alexander Kuzmenkov
4937159f85
Merge pull request #14990 from ClickHouse/aku/longer-perf 
If perf test definition changed, run everything + longer (as in master)
 2020-09-22 13:25:49 +03:00
Alexander Kuzmenkov
e646aa306f
Update columns_hashing.xml 2020-09-22 13:25:16 +03:00
Alexander Kuzmenkov
ba12e68d70
Update insert_values_with_expressions.xml 2020-09-22 13:23:10 +03:00
Alexander Kuzmenkov
4dae587e2d fix test 2020-09-21 15:02:34 +03:00
Alexander Kuzmenkov
85be1f1685 Merge remote-tracking branch 'origin/master' into HEAD 2020-09-21 14:57:25 +03:00
alexey-milovidov
a4337635cc
Update single_fixed_string_groupby.xml 2020-09-20 19:42:11 +03:00
philip.han
2045bc9f8c Fix performance_test for bloom_filter 2020-09-20 19:06:46 +09:00
Amos Bird
8ca4d6634e
better fixed string group by support 2020-09-20 15:20:09 +08:00
Alexander Kuzmenkov
e7b9319e12 If perf test definition changed, run everything + longer (as in master) 
Also some other perf test fixes
 2020-09-18 16:28:46 +03:00
philip.han
cbedd44a60 Fix indexOf() to use BloomFilter 2020-09-18 17:43:07 +09:00
Alexander Kuzmenkov
36538ce08f Don't account for short queries, we'll deal with them separately. 
New query:

```
WITH ceil(max(q[3]), 1) AS h
SELECT concat('sed -i s\'/^<test.*$/<test max_ignored_relative_change="', toString(h), '">/g\' tests/performance/', test, '.xml') AS s
FROM
(
    SELECT
        test,
        query_index,
        count(*),
        min(event_time),
        max(event_time) AS t,
        arrayMap(x -> floor(x, 3), quantiles(0, 0.5, 0.95, 1)(stat_threshold)) AS q,
        median(stat_threshold) AS m
    FROM perftest.query_metrics
    WHERE (metric = 'client_time') AND (abs(diff) < 0.05) AND (old > 0.2)
    GROUP BY
        test,
        query_index,
        query_display_name
    HAVING (t > '2020-09-01 00:00:00') AND (m > 0.1)
    ORDER BY test DESC
)
GROUP BY test
ORDER BY h DESC
FORMAT PrettySpace
```
 2020-09-17 13:00:51 +03:00
Alexander Kuzmenkov
0f8aec59a3 Adjust ignore thresholds for unstable perf tests 
Based on historical data.

```
SELECT
    test,
    ceil(max(q[3]), 1) AS h
FROM
(
    SELECT
        test,
        query_index,
        count(*),
        min(event_time),
        max(event_time) AS t,
        arrayMap(x -> floor(x, 3), quantiles(0, 0.5, 0.95, 1)(stat_threshold)) AS q,
        median(stat_threshold) AS m
    FROM perftest.query_metrics
    WHERE (metric = 'client_time') AND (abs(diff) < 0.05)
    GROUP BY
        test,
        query_index,
        query_display_name
    HAVING (t > '2020-09-01 00:00:00') AND (m > 0.1)
    ORDER BY m DESC
)
GROUP BY test
ORDER BY h DESC
FORMAT TSV

cryptographic_hashes	1.3
collations	0.8
joins_in_memory_pmj	0.8
joins_in_memory	0.7
merge_tree_simple_select	0.7
set_index	0.7
decimal_casts	0.7
website	0.6
logical_functions_medium	0.5
count	0.5
merge_tree_many_partitions	0.5
decimal_aggregates	0.5
codecs_int_insert	0.5
column_column_comparison	0.5
insert_parallel	0.4
parse_engine_file	0.4
read_in_order_many_parts	0.4
logical_functions_small	0.4
parallel_insert	0.3
parallel_index	0.3
push_down_limit	0.3
jit_large_requests	0.3
select_format	0.3
arithmetic	0.3
merge_tree_huge_pk	0.3
materialized_view_parallel_insert	0.3
columns_hashing	0.3
if_array_string	0.3
random_string	0.2
random_printable_ascii	0.2
set	0.2
empty_string_serialization	0.2
```

To apply:
```
sed 's/^\(.*\)        \(.*\)$/sed -i "s\/^<test.*$\/<test max_ignored_relative_change="'"'"\2">\/g" tests\/performance\/\1.xml/g' ../bad.tsv | bash
```
 2020-09-16 18:27:51 +03:00
Alexander Kuzmenkov
9e3797b252
Merge pull request #14710 from ClickHouse/aku/faster-perf 
Do less runs in perf test
 2020-09-14 19:17:34 +03:00