Commit Graph

68 Commits

Author SHA1 Message Date
4a474420c8 [feature](function) Add ntile function (#9867)
Add ntile function.
For non-vectorized-engine, I just implemented like Impala, rewrite ntile to row_number and count.
But for vectorized-engine, I implemented WindowFunctionNTile.
2022-06-10 10:32:40 +08:00
19bc14cf8d [feature-wip](array-type) Add array type support for vectorized parquet-orc scanner (#9856)
Only support one level array now.
for example:
- nullable(array(nullable(tinyint))) is **support**.
- nullable(array(nullable(array(xx))) is **not support**.
2022-06-09 12:11:47 +08:00
94089b9192 [Refactor] Use file factory to replace create file reader/writer (#9505)
1. Simplify code logic and improve abstraction
2. Fix the mem leak of raw pointer

Co-authored-by: lihaopeng <lihaopeng@baidu.com>
2022-06-08 15:07:39 +08:00
35c3e4e33c [Bug] runtime filter is not used as expected (#10001)
* [Bug] runtime filter is not used as expected

* update
2022-06-08 11:10:39 +08:00
35f99faa0a [Bug][Vectorized] fix core dump on vcase_expr::close (#9893)
Co-authored-by: lihaopeng <lihaopeng@baidu.com>
2022-06-01 08:05:09 +08:00
Pxl
d34d631519 [bugfix]fix TableFunctionNode memory leak (#9853) 2022-05-31 19:20:22 +08:00
c8d303a82c [bugfix] Fix BE core about vectorized join build thread memtracker switch, and FileStat duplicate 2022-05-31 19:12:42 +08:00
7199102d7c [Opt][VecLoad] Opt the vec stream load performance (#9772)
Co-authored-by: lihaopeng <lihaopeng@baidu.com>
2022-05-31 11:53:32 +08:00
f377c26bf7 [refactor][be] Optimize headers (#9708) 2022-05-30 16:12:10 +08:00
4af2493c42 [Improvement] optimize scannode concurrency query performance in vectorized engine. (#9792) 2022-05-30 16:04:40 +08:00
0683181fef [API changed](parser) Remove merge join syntax (#9795)
Remove merge join sql and merge join node
2022-05-30 09:04:21 +08:00
63aab5ee5d [Bugfix(Vec)] Fix some memory leak issues (#9824) 2022-05-29 23:04:11 +08:00
cbbda7857b [feature-wip](parquet-orc) Support orc scanner in vectorized engine (#9541) 2022-05-26 21:39:12 +08:00
f4dd3bf013 [bugfix] fix memleak in olapscannode(#9736) 2022-05-26 15:06:54 +08:00
90e8cda5f2 [Enhancement](Vectorized)build hash table with new thread, as non-vec… (#9290)
* [Enhancement][Vectorized]build hash table with new thread, as non-vectorized past do

edit after comments

* format code with clang format

Co-authored-by: lidongyang <dongyang.li@rateup.com.cn>
Co-authored-by: stephen <hello-stephen@qq.com>
2022-05-24 10:23:15 +08:00
5039ec4570 [vec][opt] opt hash join build resize hash table before insert data (#9735)
Co-authored-by: lihaopeng <lihaopeng@baidu.com>
2022-05-23 15:13:57 +08:00
500c36717d [Bug-Fix][Vectorized] Full join return error result (#9690)
Co-authored-by: lihaopeng <lihaopeng@baidu.com>
2022-05-23 13:29:37 +08:00
8fa677b59c [Refactor][Bug-Fix][Load Vec] Refactor code of basescanner and vjson/vparquet/vbroker scanner (#9666)
* [Refactor][Bug-Fix][Load Vec] Refactor code of basescanner and vjson/vparquet/vbroker scanner
1. fix bug of vjson scanner not support `range_from_file_path`
2. fix bug of vjson/vbrocker scanner core dump by src/dest slot nullable is different
3. fix bug of vparquest filter_block reference of column in not 1
4. refactor code to simple all the code

It only changed vectorized load, not original row based load.

Co-authored-by: lihaopeng <lihaopeng@baidu.com>
2022-05-20 11:43:03 +08:00
bfb1ab059d [BUG] fix information_schema.columns results not correctly on vec engine (#9612)
* VSchemaScanNode get_next bugfix

* add regression-test case for VSchemaScanNode

Co-authored-by: cambyzju <zhuxiaoli01@baidu.com>
2022-05-18 07:44:32 +08:00
Pxl
26353ba8b5 [clang build]fix clang compile error (#9615) 2022-05-18 07:42:31 +08:00
bee5c2f8aa [feature-wip](parquet-vec) Support parquet scanner in vectorized engine (#9433) 2022-05-17 09:37:17 +08:00
5660815dc6 [chore] Fix compilation errors reported by clang (#9584) 2022-05-16 22:36:16 +08:00
b817efd652 [feature] add vectorized vjson_scanner (#9311)
This pr is used to add the vectorized vjson_scanner, which can support vectorized json import in stream load flow.
2022-05-14 09:50:05 +08:00
650e3a6ba0 [feature-wip](array-type) array_contains support more nested data types (#9170)
Co-authored-by: cambyzju <zhuxiaoli01@baidu.com>
2022-05-13 12:42:40 +08:00
718a51a388 [refactor][style] Use clang-format to sort includes (#9483) 2022-05-10 21:25:35 +08:00
eec1dfde3a [feature] (vec) instead of converting line to src tuple for stream load in vectorized. (#9314)
Co-authored-by: xiepengcheng01 <xiepengcheng01@xafj-palo-rpm64.xafj.baidu.com>
2022-05-09 11:24:07 +08:00
832338c55e [improvement] set name for scanner threads and fix compile error in clang (#9336) 2022-05-05 09:53:43 +08:00
Pxl
d2374dbd5e [fix](Lateral-View) fix outer combinator not work on non-vectorized (#9212) 2022-05-01 22:09:50 +08:00
c9961c9bb9 [style] clang-format all c++ code (#9305)
- sh build-support/clang-format.sh  to  clang-format all c++ code
2022-04-29 16:14:22 +08:00
d330bc3806 [Vectorized](stream-load-vec) Support stream load in vectorized engine (#8709) (#9280)
Implement vectorized stream load.
Added fe configuration option `enable_vectorized_load` to enable vectorized stream load.

    Co-authored-by: tengjp@outlook.com
    Co-authored-by: mrhhsg@gmail.com
    Co-authored-by: minghong.zhou@163.com
    Co-authored-by: HappenLee <happenlee@hotmail.com>
    Co-authored-by: zhoubintao <35688959+zbtzbtzbt@users.noreply.github.com>
2022-04-29 09:50:51 +08:00
26bc462e1c [feature-wip] (memory tracker) (step5) Fix track bthread, fix track vectorized query (#9145)
1. fix track bthread
- Bthread, a high performance M:N thread library used by brpc. In Doris, a brpc server response runs on one bthread, possibly on multiple pthreads. Currently, MemTracker consumption relies on pthread local variables (TLS).
- This caused pthread TLS MemTracker confusion when switching pthread TLS MemTracker in brpc server response. So replacing pthread TLS with bthread TLS in the brpc server response saves the MemTracker.
Ref: 731730da85/docs/en/server.md (bthread-local)

2. fix track vectorized query
- Added track mmap. Currently, mmap allocates memory in many places of the vectorized execution engine.
- Refactored ThreadContext to avoid dependency conflicts and make it easier to debug.
- Fix some bugs.
2022-04-27 20:34:02 +08:00
Pxl
dda7604e16 [Bug][Storage-vectorized] fix code dump on outer join with not nullable column (#9112) 2022-04-21 11:02:04 +08:00
Pxl
681f960257 [fix](storage)(vectorized) query get wrong result when read datetime type column (#8872) 2022-04-18 19:34:06 +08:00
e5e0dc421d [refactor] Change ALL OLAPStatus to Status (#8855)
Currently, there are 2 status code in BE, one is common/Status.h,
and the other is olap/olap_define.h called OLAPStatus.
OLAPStatus is just an enum type, it is very simple and could not save many informations,
I will unify these code to common/Status.
2022-04-14 11:43:49 +08:00
0c8ea8ce9f [Vectorizd] Let VAssertRowNumNode handle return value of child->get_next (#8969) 2022-04-12 19:56:03 +08:00
66d2f4e1fd [fix][mem tracker] Fix MemTracker null pointer in vectorized (#8925)
Fix ThreadMemTrackerMgr::update_tracker null pointer and some details.

Issue Number: close #8920
2022-04-12 10:17:10 +08:00
c5718928df [feature-wip](array-type) support explode and explode_outer table function (#8766)
explode(ArrayColumn) desc:
> Create a row for each element in the array column. 

explode_outer(ArrayColumn) desc:
> Create a row for each element in the array column. Unlike explode, if the array is null or empty, it returns null.

Usage example:
1. create a table with array column, and insert some data;
2. open enable_lateral_view and enable_vectorized_engine;
```
set enable_lateral_view = true;
set enable_vectorized_engine=true;
```
3. use explode_outer
```
> select * from array_test;
+------+------+--------+
| k1   | k2   | k3     |
+------+------+--------+
|    3 | NULL | NULL   |
|    1 |    2 | [1, 2] |
|    2 |    3 | NULL   |
|    4 | NULL | []     |
+------+------+--------+

> select k1,explode_column from array_test LATERAL VIEW explode_outer(k3) TempExplodeView as explode_column;
+------+----------------+
| k1   | explode_column |
+------+----------------+
|    1 |              1 |
|    1 |              2 |
|    2 |           NULL |
|    4 |           NULL |
|    3 |           NULL |
+------+----------------+
```
4. explode usage example. explode return empty rows while the ARRAY is null or empty
```
> select k1,explode_column from array_test LATERAL VIEW explode(k3) TempExplodeView as explode_column;
+------+----------------+
| k1   | explode_column |
+------+----------------+
|    1 |              1 |
|    1 |              2 |
+------+----------------+
```
2022-04-08 12:11:04 +08:00
519305cb22 [feature-wip] (memory tracker) (step4) Switch TLS mem tracker to separate more detailed memory usage (#8669)
Based on #8605, Separate out the memory usage of each operator from the Query/Load/StorageEngine mem tracker.
2022-04-08 09:02:26 +08:00
98cab78320 [refactor](schema_hash) remove schema_hash since every tablet id in be is unique (#8574) 2022-04-07 08:37:45 +08:00
3698176c40 use row_size as name of variable indicating rows rather than column_size (#8803)
use row_size as name of variable indicating rows rather than column_size
2022-04-02 10:38:16 +08:00
c31c6ae91a [improvement](storage) Add more detailed timer on SegmentIter in profile (#8768)
* [improvement](storage) Add more detailed timer on SegmentIter in profile

* add OutputColumnTime
2022-04-02 10:35:28 +08:00
71ac86b183 [improvement](join) Support join project in query engine (#8722) 2022-03-31 23:00:07 +08:00
aaaaae53b5 [feature] (memory) Switch TLS mem tracker to separate more detailed memory usage (#8605)
In pr #8476, all memory usage of a process is recorded in the process mem tracker,
and all memory usage of a query is recorded in the query mem tracker,
and it is still necessary to manually call `transfer to` to track the cached memory size.

We hope to separate out more detailed memory usage based on Hook TCMalloc new/delete + TLS mem tracker.

In this pr, the more detailed mem tracker is switched to TLS, which automatically and accurately
counts more detailed memory usage than before.
2022-03-24 14:29:34 +08:00
36c85d2f06 [fix][vectorized] Fix bug of left semi/anti with other join conjunct (#8596) 2022-03-23 10:34:47 +08:00
eeae516e37 [Feature](Memory) Hook TCMalloc new/delete automatically counts to MemTracker (#8476)
Early Design Documentation: https://shimo.im/docs/DT6JXDRkdTvdyV3G

Implement a new way of memory statistics based on TCMalloc New/Delete Hook,
MemTracker and TLS, and it is expected that all memory new/delete/malloc/free
of the BE process can be counted.
2022-03-20 23:06:54 +08:00
Pxl
a824c3e489 [feature](vectorized) support lateral view (#8448) 2022-03-17 10:04:24 +08:00
e807e8b108 [improvement](memory) fix olap table scan and sink memory usage problem (#8451)
Due to unlimited queue in OlapScanNode and NodeChannel, memory usage can be
very large for reading and writing large table, e.g 'insert into tableB select * from tableA'.
2022-03-13 22:12:15 +08:00
705989d239 [improvement](VHashJoin) add probe timer (#8233) 2022-03-13 20:54:44 +08:00
e17aef9467 [refactor] refactor the implement of MemTracker, and related usage (#8322)
Modify the implementation of MemTracker:
1. Simplify a lot of useless logic;
2. Added MemTrackerTaskPool, as the ancestor of all query and import trackers, This is used to track the local memory usage of all tasks executing;
3. Add cosume/release cache, trigger a cosume/release when the memory accumulation exceeds the parameter mem_tracker_consume_min_size_bytes;
4. Add a new memory leak detection mode (Experimental feature), throw an exception when the remaining statistical value is greater than the specified range when the MemTracker is destructed, and print the accurate statistical value in HTTP, the parameter memory_leak_detection
5. Added Virtual MemTracker, cosume/release will not sync to parent. It will be used when introducing TCMalloc Hook to record memory later, to record the specified memory independently;
6. Modify the GC logic, register the buffer cached in DiskIoMgr as a GC function, and add other GC functions later;
7. Change the global root node from Root MemTracker to Process MemTracker, and remove Process MemTracker in exec_env;
8. Modify the macro that detects whether the memory has reached the upper limit, modify the parameters and default behavior of creating MemTracker, modify the error message format in mem_limit_exceeded, extend and apply transfer_to, remove Metric in MemTracker, etc.;

Modify where MemTracker is used:
1. MemPool adds a constructor to create a temporary tracker to avoid a lot of redundant code;
2. Added trackers for global objects such as ChunkAllocator and StorageEngine;
3. Added more fine-grained trackers such as ExprContext;
4. RuntimeState removes FragmentMemTracker, that is, PlanFragmentExecutor mem_tracker, which was previously used for independent statistical scan process memory, and replaces it with _scanner_mem_tracker in OlapScanNode;
5. MemTracker is no longer recorded in ReservationTracker, and ReservationTracker will be removed later;
2022-03-11 22:04:23 +08:00
68dd799796 [improvement](vectorized) Support function tuple is null (#8442) 2022-03-11 16:54:37 +08:00